Rwanda Baseline Demographic and Soil Summary and Analysis

Objective: Summary of the demographics and soil characteristics of the Rwanda long term soil health study.

Study methodology

Add in details and links on study methodology here.

library(knitr)
library(ggplot2)
library(stringr)
suppressMessages(library(dplyr))
library(sp)
suppressMessages(library(rgdal))
suppressMessages(library(dismo))
suppressMessages(library(stargazer))
library(leaflet)
library(XML)
suppressMessages(library(maptools))
library(automap)
suppressMessages(library(RStata))
suppressMessages(library(fields))
library(gstat)
library(htmltools)
suppressMessages(library(Matching))
library(reshape2)
options("RStata.StataVersion" = 12)
options("RStata.StataPath" = "/Applications/Stata/StataSE.app/Contents/MacOS/stata-se")
#chooseStataBin("/Applications/Stata/StataSE.app/Contents/MacOS/stata-se")

wd <- "/Users/mlowes/drive/optimized_agronomy/soil/soil_health_study/data/rw baseline"
dd <- paste(wd, "data", sep = "/")
od <- paste(wd, "output", sep="/")
md <- paste(wd, "maps", sep="/")
drive <- "~/drive/r_help/4_output/statistical_test_outputs"

#load data:
# This data is being drawn from the Soil lab repository. It has the baseline data with it
# d <- read.csv(paste(dd, "Rwanda_shs_commcare_soil_data_final.csv", sep="/"),  stringsAsFactors=FALSE)

Combine survey and soil data

Replicate Alex’s and Emmanuel’s merge process using “Identifiers with SSN_final” provided by Emmanuel. I use the RStata package found here

# I'm replicating Alex's do file here.
stata("merge_shs.do")

## . * merge raw commcare data with soil database
## . 
## . * date: 11 july 2016
## . 
## . clear all
## . set more off
## . 
## . * set directory
## . global wd "/Users/mlowes/drive/soil health study/data/rw baseline"
## . global dd "/Users/mlowes/drive/soil health study/data/rw baseline/data"
## . global troubleshoot "/Users/mlowes/drive/soil health study/data/rw baseline/t
## > roubleshoot"
## . 
## . * insheet data
## . insheet using "$dd/raw_rwanda_commcare_shs.csv", clear
## file /Users/mlowes/drive/soil health study/data/rw
##     baseline/data/raw_rwanda_commcare_shs.csv not found
## r(601);
## . 
## . * clean sample_id variable
## . replace sample_id = "" if sample_id == "---"
## . replace sample_id = "-99" if sample_id == "99"
## . 
## . destring sample_id, replace
## . 
## . * manipulate current "sample_id" to become a proper alpha-numeric unique iden
## > tifier
## . * this simply involves adding "C" to each numeric code that belongs to a cont
## > rol farmer
## . 
## . * variables of interest:
## .       * d_client
## .       * sample_id
## . 
## . ren demographic_infod_client d_client
## . replace d_client = "" if d_client == "---"
## . destring d_client, replace
## . 
## . tostring sample_id, gen(sample_id_string)
## . 
## . replace sample_id_string = sample_id_string + "C" if d_client == 0
## . drop sample_id 
## . ren sample_id_string sample_id
## . 
## . ***** !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ***
## > ***
## . * quantify and note cases where sample_id appears more than twice
## .      * 39 codes appear 2x
## .      * 1 code appears 3x
## .      * 1 code appears 4x
## .      * 1 code appears 10x
## .      * 1 code appears 19x
## . 
## . duplicates report sample_id
## . duplicates tag sample_id, gen(n_duplicate_id)
## . gen d_id_problem = 1 if n_duplicate_id != 0
## . replace d_id_problem = 0 if missing(d_id_problem)
## . 
## . ** need to investigate 5% of observations (114 samples), i.e. d_id_problem = 
## > 1**
## . ***** !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ***
## > ***
## . 
## . * outsheet problematic observations and examine 1-by-1
## . sort sample_id
## . outsheet demographicnom_enqueteur district demographic_infocellule_selected d
## > emographic_infoumudugudu d_client demographic_infonom_cultivateur sample_id d
## > emographic_infosex demographic_infoage_cultivateur demographic_infon_telephon
## > e_cult d_id_problem using "$dd/shs_to_check.csv" if d_id_problem == 1, replac
## > e comma
## . 
## . **************** post-investigation correction of incorrect ids *************
## > ***
## . 
## . ren demographic_infonom_cultivateur farmer_name
## . ren demographic_infocellule_selected selected_cell
## . ren demographicnom_enqueteur enumerator_name
## . 
## . * drop test/accidentally double-entered observations
## . drop if infoformid == "f138897f-6e80-4b0a-9db3-ab2134cd9e51"
## . drop if farmer_name == "Mugisha"
## . drop if farmer_name == "Muhunde yoramu"
## . drop if farmer_name == "Test"
## . drop if farmer_name == "Billy"
## . drop if farmer_name == "Gakuba michel"
## . drop if farmer_name == "Renata"
## . drop if farmer_name == "Bub"
## . drop if farmer_name == "M"
## . drop if farmer_name == "Anita"
## . drop if farmer_name == "Jean"
## . drop if farmer_name == "Nyirazaninka lea"
## . drop if farmer_name == "Mugiraneza Pacifiue"
## . drop if farmer_name == "Nsengiyaremye Innocent"
## . 
## . * correct incorrectly recorded ids
## . replace sample_id = "880C" if farmer_name == "Nyiransanzineza Marie Rose"
## . replace sample_id = "811C" if farmer_name == "Ngirabatware  Daniel"
## . replace sample_id = "646C" if farmer_name == "Nyirasabwa anonciata"
## . replace sample_id = "624C" if farmer_name == "Nyirinkindi j Bosco"
## . replace sample_id = "560C" if farmer_name == "Mubirigi Bertin"
## . replace d_client = 0 if farmer_name == "Mubirigi Bertin"
## . replace sample_id = "375C" if farmer_name == "Hitimana ezecquiere"
## . replace sample_id = "322C" if farmer_name == "Nyirantabire arrivera"
## . replace sample_id = "2566C" if farmer_name == "MUSABYIMANA Colletta"
## . replace sample_id = "2399C" if farmer_name == "Nyirabajyambere anastisia"
## . replace d_client = 0 if farmer_name == "Nyirabajyambere anastisia"
## . replace sample_id = "2280C" if farmer_name == "Zihabake Simon"
## . replace sample_id = "2202C" if farmer_name == "mugiraneza pacifique"
## . replace sample_id = "1780C" if farmer_name == "Hakuzimana Theophile"
## . replace sample_id = "1757C" if farmer_name == "Bavugirije Feleciane"
## . replace sample_id = "1626C" if farmer_name == "Nyabivumu"
## . replace sample_id = "1575C" if farmer_name == "BAHIMBANDE Beriyana."
## . replace d_client = 0 if farmer_name == "BAHIMBANDE Beriyana."
## . replace sample_id = "1103C" if farmer_name == "Barirwanda Elyse"
## . replace sample_id = "1037C" if farmer_name == "Mukashema Faraziya"
## . replace d_client = 0 if farmer_name == "Mukashema Faraziya" 
## . replace sample_id = "3069" if farmer_name == "Ndababonye Silas"
## . replace sample_id = "2611" if farmer_name == "Sebazungu modeste"
## . replace d_client = 1 if farmer_name == "Sebazungu modeste"
## . replace sample_id = "2415" if farmer_name == "Nkaka joseph"
## . replace sample_id = "2412" if farmer_name == "Ndagijimana alexis"
## . replace sample_id = "2410" if farmer_name == "Sebuhoro elie"
## . replace sample_id = "2406" if farmer_name == "Mugisha ruth"
## . replace sample_id = "2405" if farmer_name == "Mboneza hasan"
## . replace sample_id = "2404" if farmer_name == "Ntibitonderwa veriane"
## . replace sample_id = "2399" if farmer_name == "Nyiraburakeye feresita"
## . replace sample_id = "2395" if farmer_name == "Uwimana danier"
## . replace sample_id = "2394" if farmer_name == "Bizinde silas"
## . replace sample_id = "2390" if farmer_name == "Nyirahabimvana keresesiya"
## . replace sample_id = "2388" if farmer_name == "Nyirasekerabanzi tharcilla"
## . replace sample_id = "2386" if farmer_name == "Ndererimana emmanuel"
## . replace sample_id = "2291" if farmer_name == "Ndacyayisenga Feresiyani"
## . replace sample_id = "2184" if farmer_name == "Muhawenimana Beretirida"
## . replace sample_id = "2145" if farmer_name == "Fapfakwita celestine"
## . replace sample_id = "2141" if farmer_name == "Nsangiranabo krizoroji"
## . replace sample_id = "2140" if farmer_name == "Nsabyumuremyi anakereti"
## . replace sample_id = "2136" if farmer_name == "Nyirantibitonderwa savoroniya"
## . replace sample_id = "2131" if farmer_name == "Hagenimana Ewufroni"
## . replace sample_id = "2066" if farmer_name == "Hitayezu vicent"
## . replace d_client = 1 if farmer_name == "Hitayezu vicent"
## . replace sample_id = "2044" if farmer_name == "Niyodushima janette"
## . replace d_client = 1 if farmer_name == "Niyodushima janette"
## . replace sample_id = "2024" if farmer_name == "Mukabatsinda veren A"
## . replace sample_id = "2020" if farmer_name == "Sibomana innocent"
## . replace sample_id = "2009" if farmer_name == "Nsabimana inyasi"
## . replace sample_id = "2002" if farmer_name == "Mateso elizabet"
## . replace sample_id = "1889" if farmer_name == "Niyonsaba Daniel"
## . replace sample_id = "1864" if farmer_name == "Mukankubana Souzanne"
## . replace sample_id = "1780" if farmer_name == "Nyirantihabose Annonciatha"
## . replace sample_id = "1771" if farmer_name == "Bizimungu Damascne"
## . replace sample_id = "1675" if farmer_name == "Bikorimana Isaac"
## . replace sample_id = "1103" if farmer_name == "Munyabarata japhette"
## . replace sample_id = "954" if farmer_name == "Hahirwabake silver"
## . replace sample_id = "541" if farmer_name == "Rugemintwaza Vianney"
## . replace sample_id = "454" if farmer_name == "Niyonizeye fororida"
## . replace sample_id = "407" if farmer_name == "Munyaneza jean pierre"
## . replace sample_id = "375" if farmer_name == "Munyarubuga nanias"
## . replace sample_id = "322" if farmer_name == "Gacandaga zackarie"
## . replace sample_id = "262" if farmer_name == "MUHIRWA J Damascene"
## . replace sample_id = "32" if farmer_name == "Nsanzemuhire"
## . 
## . *** re-run duplicates test
## . drop d_id_problem
## . drop n_duplicate_id
## . 
## . duplicates report sample_id
## . duplicates tag sample_id, gen(n_duplicate_id)
## . gen d_id_problem = 1 if n_duplicate_id != 0
## . replace d_id_problem = 0 if missing(d_id_problem)
## . 
## . * second try: outsheet problematic observations and examine 1-by-1
## . sort sample_id
## . outsheet using "$dd/shs_to_check_2.csv" if d_id_problem == 1, replace comma
## . 
## . * second try: correct incorrect ids
## . 
## . replace sample_id = "2202" if farmer_name == "Mukamurara scholastique"
## . replace d_client = 1 if farmer_name == "Mukamurara scholastique"
## . replace sample_id = "2195" if farmer_name == "Ndacyayisenga Feresiyani"
## . replace sample_id = "824" if farmer_name == "Ngirabatware  Daniel"
## . 
## . replace sample_id = "" if rownumber == 238
## . replace sample_id = "" if rownumber == 848
## . replace sample_id = "" if rownumber == 1257
## . replace sample_id = "" if rownumber == 1732
## . 
## . *** re-run duplicates test
## . drop d_id_problem
## . drop n_duplicate_id
## . 
## . duplicates report sample_id
## . duplicates tag sample_id, gen(n_duplicate_id)
## . gen d_id_problem = 1 if n_duplicate_id != 0
## . replace d_id_problem = 0 if missing(d_id_problem)
## . 
## . * third try: outsheet problematic observations
## . sort sample_id
## . outsheet using "$dd/shs_to_check_3.csv" if d_id_problem == 1, replace comma
## . 
## . *** 4 sets of duplicates remain (2 observations)
## . drop if d_id_problem == 1
## . save "$dd/rwanda_commcare_shs_clean.dta", replace
## . outsheet using "$dd/rwanda_commcare_shs_clean.csv", comma replace
## . 
## . * merge datasets
## . insheet using "$dd/mne_rwanda_database_shs.csv", clear
## . 
## . save "$dd/mne_rwanda_database_shs.dta", replace
## . 
## . use "$dd/rwanda_commcare_shs_clean.dta", clear
## . 
## . merge 1:1 sample_id using "$dd/mne_rwanda_database_shs.dta"
## . 
## .    * 93 observations unmatched... :(
## . 
## . * investigate unmatched observations
## . sort sample_id
## . outsheet using "$troubleshoot/unmatched_commcare.csv" if _merge == 1, comma r
## > eplace
## . outsheet using "$troubleshoot/unmatched_inventory.csv" if _merge == 2, comma 
## > replace
## . 
## . * make first round of corrections in master data
## . use "$dd/rwanda_commcare_shs_clean.dta", clear
## . 
## . replace sample_id = "2711C" if farmer_name == "Kamuhanda Elie"
## . replace sample_id = "260C" if farmer_name == "HATEGEKIMANA Mathias"
## . replace sample_id = "245" if farmer_name == "Kayiranga Michel"
## . replace sample_id = "1543" if farmer_name == "Ndagijimana Eliezel"
## . replace sample_id = "1543C" if farmer_name == "Mukantanganzwa odeta"
## . replace sample_id = "1632" if farmer_name == "Nyiranziguye Cecile"
## . replace sample_id = "2782" if farmer_name == "Mukamungu leocadie"
## . replace sample_id = "421C" if farmer_name == "Niyonizeye Francine"
## . replace sample_id = "828C" if farmer_name == "Mukamazimpaka Terese"
## . replace sample_id = "364" if farmer_name == "Mukamuyango verina"
## . replace sample_id = "364C" if farmer_name == "Mujawimana jeanne"
## . replace sample_id = "368" if farmer_name == "Nyiramanzi jean damascene"
## . replace sample_id = "368C" if farmer_name == "Karwiyegura rachel"
## . replace sample_id = "391" if farmer_name == "Mugemane augistin"
## . replace sample_id = "391C" if farmer_name == "Ngirishuti sumayire"
## . replace sample_id = "395" if farmer_name == "Niyomugabo amiel"
## . replace sample_id = "395C" if farmer_name == "Ntawiniga augustin"
## . replace sample_id = "396" if farmer_name == "Mpayimana philippe"
## . replace sample_id = "329" if farmer_name == "Mucyezangango emmanuel"
## . replace sample_id = "411" if farmer_name == "Nsabimana mathias"
## . replace sample_id = "411C" if farmer_name == "Singirankabo"
## . replace sample_id = "414" if farmer_name == "Rutaharama eldephonse"
## . replace sample_id = "329C" if farmer_name == "Mukamana josephine"
## . replace sample_id = "657" if farmer_name == "Nyandwi vincent"
## . replace sample_id = "569C" if farmer_name == "Uwimana Bercilla"
## . replace sample_id = "2051C" if farmer_name == "Ntambabazi Anastase"
## . replace sample_id = "2249" if farmer_name == "Uwineza valentine"
## . replace sample_id = "2360C" if farmer_name == "Bamporineza j deDieu"
## . replace sample_id = "2001" if farmer_name == "Nyiramakuba thanene"
## . replace sample_id = "2897C" if farmer_name == "Mukagatanazi Marianne"
## . replace sample_id = "2965C" if farmer_name == "Twizeyimana Theoneste"
## . 
## . drop if farmer_name == "Yjk"
## . drop if farmer_name == "Uwambajimana Agnes"
## . drop if farmer_name == "Tr"
## . 
## . replace sample_id = "2415C" if farmer_name == "Nyiramahirwe Frolance"
## . replace sample_id = "851" if farmer_name == "Bigirimana  Amon"
## . 
## . 
## . * duplicates check
## . drop d_id_problem
## . drop n_duplicate_id
## . 
## . duplicates report sample_id
## . duplicates tag sample_id, gen(n_duplicate_id)
## . gen d_id_problem = 1 if n_duplicate_id != 0
## . replace d_id_problem = 0 if missing(d_id_problem)
## . 
## . * second try: merge
## . merge 1:1 sample_id using "$dd/mne_rwanda_database_shs.dta"
## . 
## . * deal with 30 unmatched cases
## . sort sample_id
## . outsheet using "$troubleshoot/unmatched_commcare_2.csv" if _merge == 1, comma
## >  replace
## . outsheet using "$troubleshoot/unmatched_inventory_2.csv" if _merge == 2, comm
## > a replace
## . 
## . * 5 submitted surveys with no corresponding sample
## . *drop if _merge == 1
## . 
## . * 19 samples with no corresponding survey
## . *drop if _merge == 2
## . 
## . * outsheet merged database
## . outsheet using "$dd/rwanda_shs_baseline_data.csv", comma replace
## . * THIS IS THE CLEAN VERSION! ONLY USE THIS! DON'T USE OTHER THINGS!
## .

Import the results of Alex’s do file.

d <- read.csv(paste(dd, "rwanda_shs_baseline_data.csv", sep = "/"), stringsAsFactors = F)
Identifiers <- read.csv(paste(dd, "Identifiers with SSN_final.csv", sep = "/"), stringsAsFactors = F)

wetChem <- read.csv(paste("/Users/mlowes/drive/JuPyteR/robert.on@oneacrefund.org/Rwanda Acidity", "Original chem_rwanda_shs.csv", sep = "/"))


d <- left_join(d, Identifiers, by="sample_id")

# now import the soil results and merge with survey data
soilRF <- read.csv(paste(dd, "rwshs_rfresults.csv", sep = "/"), stringsAsFactors = F)
names(soilRF)[names(soilRF)=="X"] <- "SSN"

d <- left_join(d, soilRF, by="SSN")

Cleaning baseline variables

Now let’s start cleaning the demographic variables

d[d=="---"] <- NA

names(d) <- gsub("text_final_questions", "", names(d))
names(d) <- gsub("intro_champ_echantillon", "", names(d))
names(d) <- gsub("demographic_info", "", names(d))
names(d) <- gsub("other_inputs_", "", names(d))
names(d) <- gsub("crop1_15b_inputs", "", names(d))
names(d) <- gsub("crop2_15b_inputs", "", names(d))
names(d) <- gsub("^15b", "", names(d))
names(d) <- gsub("historical_intro", "", names(d))

names(d)[names(d)=="field_dim"] <- "field_dim1"
names(d)[names(d)=="v51"] <- "field_dim2"

Address unusual field sizes

ggplot(d, aes(x=field_dim1, y=field_dim2)) + geom_point()

## Warning: Removed 19 rows containing missing values (geom_point).

It seems really unlikely that fields are 200 meters long while only being 20 meters wide. I don’t know how to check this though.

# clean field dimensions here - winsor the values to something reasonable.
d[(d$field_dim1>=100 | d$field_dim2>=100) & !is.na(d$field_dim1), c("field_dim1", "field_dim2")]

##      field_dim1 field_dim2
## 125          85        125
## 126         200         10
## 127          20        152
## 130         140        250
## 140          50        135
## 211          12        107
## 415           5        135
## 452         108         65
## 626         100         40
## 836          12        153
## 844          10        212
## 1153        150         20
## 1154        150        160
## 1155        100         40
## 1239        102         57
## 1241        112         13
## 1314        140          8
## 1345        104         24
## 1347        123          6
## 1378        120          9
## 1388        100          7
## 1389        200          4
## 1397        160          4
## 1399        150         50
## 1420        200         20
## 1421        200         18
## 1440         30        105
## 1457         29        101
## 1476        102         73
## 1477        108         62
## 1525         12        123
## 1539        100         95
## 1567        100          8
## 1709        100          6
## 1859        157          9
## 1976         15        120
## 1979         25        102
## 2074          7        100
## 2411         30        153
## 2452        203          8

Take care of demographic data formatting issues

# deal with names and drop unnecessary variables
d <- d %>% 
    dplyr::select(-c(rownumber, infoformid, introductiond_accept, photo,
        infocompleted_time, 
        enumerator_name, contains("phone"), farmer_name, farmersurname, farmername,
        d_respondent, additionalsamplepackedandsenttol, additionalsamplerequestedfromlab,
        datedryingcompleteifnecessary, driedindistrictifnecessary, senttohqyo,
        collectedindistrictyo, excessstoredathq_, receivedathq_,dateofinitialdryingifnecessary,
        samplecollectedinfieldyo, field_des, samplewetordry)) %>%
    rename(
    female = sex,
    age = age_cultivateur,
    own = d_own,
    client = d_client) %>%
    mutate(
    female= ifelse(female=="gore", 1,0),
    field.size = field_dim1*field_dim2
    )

d$total.seasons <- apply(d[, grep("d_season_list", names(d))], 1, function(x) {
    sum(x, na.rm=T)})

Clean agronomic practice variables

agVars <- c("n_season_fert", "n_season_compost", "n_season_lime", "n_season_fallow",
            "n_seasons_leg_1", "n_seasons_leg_2")
summary(d[,agVars])

##  n_season_fert    n_season_compost n_season_lime     n_season_fallow  
##  Min.   : 0.000   Min.   : 0.000   Min.   : 0.0000   Min.   : 0.0000  
##  1st Qu.: 0.000   1st Qu.: 2.000   1st Qu.: 0.0000   1st Qu.: 0.0000  
##  Median : 1.000   Median : 5.000   Median : 0.0000   Median : 0.0000  
##  Mean   : 2.041   Mean   : 5.651   Mean   : 0.1819   Mean   : 0.6355  
##  3rd Qu.: 3.000   3rd Qu.:10.000   3rd Qu.: 0.0000   3rd Qu.: 0.0000  
##  Max.   :10.000   Max.   :10.000   Max.   :10.0000   Max.   :10.0000  
##  NA's   :19       NA's   :19       NA's   :19        NA's   :19       
##  n_seasons_leg_1  n_seasons_leg_2   
##  Min.   : 0.000   Min.   : -88.000  
##  1st Qu.: 0.000   1st Qu.:   0.000  
##  Median : 1.000   Median :   2.000  
##  Mean   : 2.171   Mean   :   4.368  
##  3rd Qu.: 4.000   3rd Qu.:   5.000  
##  Max.   :10.000   Max.   :3333.000  
##  NA's   :19       NA's   :19

Sort out the legumes as a second crop

table(d$n_seasons_leg_2, useNA = 'ifany')

## 
##  -88    0    1    2    3    4    5    6    7    8    9   10   15   22   50 
##    1  956  221  258  174  130  301   61   34   66   57  199    1    1    1 
##   53   83   88  101  102 3333 <NA> 
##    3    1    1    1    1    1   19

d$n_seasons_leg_2 <- ifelse(d$n_seasons_leg_2 <0 | d$n_seasons_leg_2>10, NA, d$n_seasons_leg_2)

Check out client variable - why are there missing values?

table(d$client, useNA = 'ifany')

## 
##    0    1 <NA> 
## 1233 1236   19

d[is.na(d$client), c("sample_id")]

##  [1] "1189C" "1207C" "1295"  "1295C" "1298C" "1300"  "1300C" "1366C"
##  [9] "1476C" "1485C" "1554"  "1573"  "1614"  "2042"  "2371C" "2373" 
## [17] "2375"  "2382"  "2442"

# replace client based on whether there is a C in the client variable.
d$client.check <- ifelse(grepl("C", d$sample_id)==T, 0, 1)
table(d$client, d$client.check)

##    
##        0    1
##   0 1233    0
##   1    1 1235

It looks like most farmers were recorded correctly except for one farmer who was coded as Tubura farmer but their sample_id indicate their a control. Let’s take a look:

d[d$client==1 & d$client.check==0 & !is.na(d$d_gps), c("sample_id", "tuburacontroltc")]

##     sample_id tuburacontroltc
## 990     2202C         Control

# they should be a control
d[d$client==1 & d$client.check==0 & !is.na(d$d_gps), "client"] <- 0

# remove farmers for which we have soil data but no survey data (using)
d <- d[-which(grepl("using", d$X_merge)),]
table(d$client, d$client.check, useNA = 'ifany')

##    
##        0    1
##   0 1234    0
##   1    0 1235

# update client to equal client check
d$client <- d$client.check

Fix some more variable names:

names(d)[names(d)=="field_kg_fert1_1"] <- "field_kg_fert1_15b"
names(d)[names(d)=="field_kg_fert2_1"] <- "field_kg_fert2_15b"
names(d)[names(d)=="field_kg_compost"] <- "field_kg_compost_15b"

Recode variables to numeric:

# recode to numeric
varlist <- c("client", "own", "crop1_15b_seedkg", "crop1_15b_yield", "crop1_15b_yield_",
    "crop2_15b_seedkg", "crop2_15b_yield", "crop2_15b_yield_", "field_kg_fert1_15b",
    "field_kg_fert2_15b", "field_kg_compost_15b", "d_lime_15b", "kg_lime_15b")

# check that there aren't values hidden in the character variables
#apply(d[,varlist], 2, function(x){table(x, useNA='ifany')})

# recode characters to numerics
d[, varlist] <- sapply(d[,varlist], as.numeric)

Sort out kgs of lime applied

table(d$kg_lime_15b, useNA = 'ifany')

## 
##  -88    1    3    4    5   10   13   15   20   25   30   35   50   60   88 
##    1    2    2    2    6    5    1    1    2    5    1    1    8    1    2 
##  100  150  200 <NA> 
##    2    4    1 2422

d$kg_lime_15b <- ifelse(abs(d$kg_lime_15b)==88, NA, d$kg_lime_15b)

# divide out GPS coordinates
# http://rfunction.com/archives/1499

# replace the blank gps_pic_guide with info
d <- cbind(d, str_split_fixed(d$gps_pic_guid, " ", n=4))
names(d)[names(d)=="1" |names(d)== "2" | names(d)== "3" | names(d)== "4"] <- c("lat", "lon", "alt", "precision")
d[,c("lat", "lon", "alt", "precision")] <- sapply(d[,c("lat", "lon", "alt", "precision")],
                                                  function(x){as.numeric(as.character(x))})

Cleaning soil data

Cleaning of soil data: Come back, check and clean the soil data before outputting to clean data set. Plot each of the soil variables to look for unrealistic values.

dim(d[is.na(d$m3.Ca),])

## [1]  26 111

d <- d[-which(is.na(d$m3.Ca)),]
       
summary(d[,c("m3.Ca", "m3.Mg", "pH", "Total.N", "Total.C", "ExAl")])

##      m3.Ca            m3.Mg               pH           Total.N       
##  Min.   : 213.8   Min.   :  36.93   Min.   :4.549   Min.   :0.05723  
##  1st Qu.: 436.5   1st Qu.: 115.94   1st Qu.:5.020   1st Qu.:0.13577  
##  Median : 723.1   Median : 184.63   Median :5.535   Median :0.15483  
##  Mean   : 877.1   Mean   : 208.23   Mean   :5.543   Mean   :0.15706  
##  3rd Qu.:1159.5   3rd Qu.: 270.55   3rd Qu.:6.010   3rd Qu.:0.17910  
##  Max.   :3724.8   Max.   :1009.88   Max.   :7.131   Max.   :0.24656  
##     Total.C            ExAl        
##  Min.   :0.9107   Min.   :0.02589  
##  1st Qu.:1.7298   1st Qu.:0.09995  
##  Median :2.1164   Median :0.23170  
##  Mean   :2.1147   Mean   :0.57795  
##  3rd Qu.:2.3664   3rd Qu.:1.00947  
##  Max.   :4.2212   Max.   :2.31905

Let’s check out the rows for which we don’t have soil data and drop them as they won’t contribute to the full picture.

Graphs of RW baseline soil variables

soilVars <- c("m3.Ca", "m3.Mg", "pH", "Total.N", "Total.C", "ExAl")

for(i in 1:length(soilVars)){
print(
  ggplot(data=d, aes(x=as.factor(client), y=d[,soilVars[i]])) + 
    geom_boxplot() +
    labs(x="Tubura Farmer", y=soilVars[i], title = paste("RW baseline soil - ", soilVars[i], sep = ""))
  )  
}

Check out soil relationships

There are biologically predictable relationships between soil chemical characteristics. For instance, we expect Ca and Mg to move in the same direction and be positively correlated with pH. If we had Aluminum as an outcome, we’d expect pH to be negatively correlated with soluable aluminum. Let’s look quickly to confirm if those relationships are present:

ggplot(d, aes(x=m3.Ca, y=m3.Mg)) + geom_point() +
    stat_smooth(method="loess") +
    labs(x = "Calcium (m3)", y= "Magnesium (m3)", title="Calcium and Magnesium relationship")

ggplot(d, aes(x=pH, y=m3.Ca)) + geom_point() +
  stat_smooth(method="loess") +
  labs(x = "pH", y="Calcium (m3)", title = "pH and Calcium relationship")

ggplot(d, aes(x=pH, y=m3.Mg)) + geom_point() +
  stat_smooth(method="loess") +
  labs(x = "pH", y="Magnesium (m3)", title = "pH and Magnesium relationship")

ggplot(d, aes(x=pH, y=ExAl)) + geom_point() +
  stat_smooth(method="loess") +
  labs(x = "pH", y="Exchangeable Aluminum", title = "pH and Aluminum relationship")

ggplot(d, aes(x=Total.C, y=Total.N)) + geom_point() + 
  stat_smooth(method="loess") +
  labs(x = "Total Carbon", y="Total Nitrogen", title = "Carbon and Nitrogen relationship")

The soil characteristics are moving in the manner that is consistent with our understanding of soil chemical processes.

Save clean demographic and soil data to external file

write.csv(d, file=paste(dd, "shs rw baseline.csv", sep = "/"))
save(d, file=paste(dd, "shs rw baseline.Rdata", sep = "/"))

Map of baseline observations

Produce a simple map of where our observations are

See here for more on using markerClusterOptions in leaflet.

In the map below, the larger green circles are Tubura farmers and the smaller blue circles are control farmers.

e <- d[!is.na(d$lon),]
ss <- SpatialPointsDataFrame(coords = e[, c("lon", "lat")], data=e)

pal <- colorNumeric(c("navy", "green"), domain=unique(ss$client))
map <- leaflet() %>% addTiles() %>%
  setView(lng=rwanda$longitude, lat=rwanda$latitude, zoom=8) %>%
  addCircleMarkers(lng=ss$lon, lat=ss$lat, 
                   radius= ifelse(ss$client==1, 10,6),
                   color = pal(ss$client),
clusterOptions = markerClusterOptions(disableClusteringAtZoom=13, spiderfyOnMaxZoom=FALSE))

map

Summary statistics

Table of final baseline breakdown

count <- d %>% group_by(district) %>% 
  dplyr::summarize(
    t.count = sum(ifelse(client==1,1,0)),
    c.count = sum(ifelse(client==0,1,0)),
    total = n()
  ) %>% ungroup()

count <- as.data.frame(count)
write.csv(count, file=paste(od, "final rw sample breakdown.csv", sep="/"), row.names=F)
as.data.frame(count)

##        district t.count c.count total
## 1   Gatsibo_LWH      58      59   117
## 2  Gatsibo_NLWH      87      81   168
## 3      Gisagara      46      46    92
## 4          Huye     114     115   229
## 5       Karongi     163     159   322
## 6       Kayonza      55      58   113
## 7      Mugonero     131     129   260
## 8     Nyamagabe      56      56   112
## 9    Nyamasheke     147     146   293
## 10       Nyanza      46      46    92
## 11    Nyaruguru      46      46    92
## 12      Rutsiro     165     167   332
## 13    Rwamagana     110     111   221

Baseline balance

Let’s see how balanced our farmers are in terms of demographic variables. Tubura farmers were selected based on (list criteria) and control farmers in the same area tha fit the same criteria were also selected. No matching process has been performed to identify the control farmers that most closely resemble the Tubura farmers in the sample. These results are entirely reflecting the balance inherent in the identification process, not any statistical matching of treatment and control.

d$valley <- ifelse(d$general_field_infofield_location=="valley", 1,0)
d$hillside <- ifelse(d$general_field_infofield_location=="hillside", 1,0)
d$hilltop <- ifelse(d$general_field_infofield_location=="hilltop", 1,0)
names(d)[names(d)=="general_field_infograde_hill"] <- "slope"

cor(d[, grep("betail_", names(d))], use='complete.obs')

##                        betail_ownedn_inka betail_ownedn_ihene
## betail_ownedn_inka             1.00000000         0.036402152
## betail_ownedn_ihene            0.03640215         1.000000000
## betail_ownedn_inkoko           0.09740274         0.127424693
## betail_ownedn_ingurube         0.02166420        -0.009667001
## betail_ownedn_intama           0.04245052         0.002625236
##                        betail_ownedn_inkoko betail_ownedn_ingurube
## betail_ownedn_inka               0.09740274            0.021664199
## betail_ownedn_ihene              0.12742469           -0.009667001
## betail_ownedn_inkoko             1.00000000            0.049416370
## betail_ownedn_ingurube           0.04941637            1.000000000
## betail_ownedn_intama             0.03041599            0.025487680
##                        betail_ownedn_intama
## betail_ownedn_inka              0.042450521
## betail_ownedn_ihene             0.002625236
## betail_ownedn_inkoko            0.030415986
## betail_ownedn_ingurube          0.025487680
## betail_ownedn_intama            1.000000000

names(d)[grep("betail_", names(d))] <- c("cows", "goats", "chickens", "pigs", "sheep")

d$own.cows <- ifelse(d$cows>0 & !is.na(d$cows), 1,0)
d$own.goats <- ifelse(d$goats>0 & !is.na(d$goats), 1,0)
d$own.chickens <- ifelse(d$chickens>0 & !is.na(d$chickens), 1,0)
d$own.pigs <- ifelse(d$pigs>0 & !is.na(d$pigs), 1,0)
d$own.sheep <- ifelse(d$sheep>0 & !is.na(d$sheep), 1,0)

Crop selection in 15B

names(d)[names(d)=="inputuse_15b_priorculture_15b_1"] <-"crop_15b"
d$crop_15b <- tolower(d$crop_15b)
sort(prop.table(table(d$crop_15b, useNA = 'ifany')))

## 
##          kor         hwag        sheke       coffee          cer 
## 0.0004093328 0.0008186656 0.0008186656 0.0012279984 0.0024559967 
##        uburo         teke       insina         shaz          nyo 
## 0.0036839951 0.0040933279 0.0057306590 0.0085959885 0.0135079820 
##          gan    other_veg          ray          yum       ntacyo 
## 0.0196479738 0.0212853050 0.0519852640 0.0577159230 0.0589439214 
##          gor         soya          jum          big         saka 
## 0.0597625870 0.0790012280 0.0920998772 0.1604584527 0.1649611134 
##          shy 
## 0.1927957429

d$climbing_bean <- ifelse(d$crop_15b=="shy", 1,0)
d$sorghum <- ifelse(d$crop_15b=="saka", 1,0)
d$bush_bean <- ifelse(d$crop_15b=="big", 1,0)
d$maize <- ifelse(d$crop_15b=="gor", 1,0)

out.list <- c("female", "age", "hhsize", "own", "field.size",
              "n_season_fert", "n_season_compost", "n_season_lime", "n_season_fallow", "n_seasons_leg_1", "n_seasons_leg_2", "slope",  "alt", "valley", "hillside", "hilltop", "cows", "goats", "chickens", "pigs", "sheep", "climbing_bean", "sorghum", "bush_bean","maize", soilVars, "own.cows", "own.goats", "own.chickens", "own.pigs", "own.sheep")

output <- do.call(rbind, lapply(out.list, function(x) {
  
  out <- t.test(d[,x] ~ d[,"client"], data=d)
  tab <- data.frame(out[[5]][[2]],out[[5]][[1]], out[3])
  tab[,1:2] <- round(tab[,1:2],3)
  names(tab) <- c(names(out[[5]]), "pvalue")
  return(tab)
}))

# use p.adjust with bonferroni correction
output$pvalue <- p.adjust(output$pvalue, method="fdr")
rownames(output) <- out.list
output <- output[order(output$pvalue),]
output$pvalue <- ifelse(output[, 3] < 0.001, "< 0.001", round(output[, 3], 3)) 


colnames(output) <- c("Tubura Client","Non-Tubura", "p-value")  

print(kable(output))

	Tubura Client	Non-Tubura	p-value
n_season_fert	3.263	0.833	< 0.001
female	0.493	0.649	< 0.001
own.cows	0.683	0.539	< 0.001
age	44.021	48.088	< 0.001
hhsize	5.418	4.894	< 0.001
own.chickens	0.342	0.267	< 0.001
n_seasons_leg_2	2.550	3.072	< 0.001
n_season_lime	0.228	0.132	< 0.001
own.pigs	0.355	0.281	< 0.001
pigs	0.525	0.394	0.006
chickens	1.225	0.895	0.006
hillside	0.682	0.733	0.016
own	0.932	0.957	0.023
cows	1.178	0.912	0.023
valley	0.266	0.222	0.025
maize	0.072	0.048	0.025
sorghum	0.146	0.184	0.026
climbing_bean	0.212	0.174	0.036
own.sheep	0.104	0.077	0.041
sheep	0.202	0.142	0.049
n_season_compost	5.818	5.523	0.095
Total.N	0.156	0.158	0.102
n_season_fallow	0.690	0.569	0.103
m3.Mg	204.339	212.129	0.171
m3.Ca	859.917	894.354	0.181
field.size	667.734	601.671	0.214
Total.C	2.104	2.126	0.393
bush_bean	0.154	0.167	0.456
n_seasons_leg_1	2.228	2.135	0.472
hilltop	0.051	0.044	0.472
pH	5.533	5.553	0.472
slope	10.203	10.445	0.547
ExAl	0.584	0.571	0.654
goats	1.076	1.053	0.768
alt	1570.267	1565.917	0.863
own.goats	0.485	0.485	0.981

#write table
write.csv(output, file=paste(od, "baseline balance.csv", sep="/"), row.names=T)

# write baseline balance table for ES per his table layouts
t4order <- c("age", "female", "hhsize", "own")
table4vars <- paste(t4order, collapse="|")

rw.table4 <- output[grepl(table4vars, rownames(output)),]
rw.table4 <- rw.table4[order(match(rownames(rw.table4), t4order)),]
write.csv(rw.table4, file=paste(od, "pre match balance table4.csv", sep="/"),
          row.names = T)

t5order <- c("field.size", "slope", "alt", "hilltop", "hillside", "valley",
                      "climbing_bean", "sorghum", "bush_bean", "maize")

table5vars <- paste(t5order,collapse="|")

rw.table5 <- output[grepl(table5vars, rownames(output)),]
rw.table5 <- rw.table5[order(match(rownames(rw.table5), t5order)),]
write.csv(rw.table5, file=paste(od, "pre match balance table5.csv", sep = "/"),
          row.names = T)

t6order <- c("own.cows", "cows", "own.pigs", "pigs", "own.sheep", "sheep", "own.goats", "goats","own.chickens", "chickens")
table6vars <- paste(t6order, collapse = "|")
rw.table6 <- output[grep(table6vars, rownames(output)), ]
rw.table6 <- rw.table6[order(match(rownames(rw.table6), t6order)),]
write.csv(rw.table6, file=paste(od, "pre match balance table6.csv", sep = "/"), 
          row.names=T)

Overall balance interpretation

Demographic variables We are not well balanced along the main demographic variables we collected, sex, age and HH size. For the purposes of inference we can test some matching algorithms to improve the match between Tubura and control farmers.

Agricultural practice variables We are decently balanced along agricultural practice variables. Our course of action here is similiar to our options with the demographic variables.

Soil Variables We are balanced on the primary soil variables of interest betwen our Tubura farmers and the comparison farmers.

Baseline balance by district

dist.output <- do.call(rbind, lapply(split(d, d$district), function(x) {
  
  tab <- do.call(rbind, lapply(out.list, function(y) {
    
    out <- t.test(x[,y] ~ x[,"client"], data=x)
    tab <- data.frame(out[[5]][[2]],out[[5]][[1]], out[3])
    tab[,1:2] <- round(tab[,1:2],3)
    names(tab) <- c(names(out[[5]]), "pvalue")
    #tab[,3] <- p.adjust(tab[,3], method="holm")
    #tab[,3] <- ifelse(tab[,3] < 0.001, "< 0.001", round(tab[,3],3))
    #print(tab)
    return(tab)
  }))
  
  return(data.frame(district = unique(x$district), tab))
}))

rownames(dist.output) <- NULL
dist.output$variable <- rep(out.list,length(unique(d$district)))    

# order variables 
dist.output <- dist.output[, c(1, 5, 2:4)]
dist.output$pvalue <- p.adjust(dist.output$pvalue, method="fdr")
dist.output <- dist.output[order(dist.output$pvalue),]

dist.output$pvalue <- ifelse(dist.output$pvalue < 0.001, "< 0.001", round(dist.output$pvalue,3))
colnames(dist.output) <- c("District", "Varible", "Tubura Client","Non-Tubura", "p-value")  

print(kable(dist.output))

	District	Varible	Tubura Client	Non-Tubura	p-value
150	Karongi	n_season_fert	3.785	0.447	< 0.001
222	Mugonero	n_season_fert	4.313	0.597	< 0.001
402	Rutsiro	n_season_fert	4.200	1.066	< 0.001
294	Nyamasheke	n_season_fert	5.068	1.507	< 0.001
438	Rwamagana	n_season_fert	2.645	1.054	< 0.001
114	Huye	n_season_fert	2.246	0.322	< 0.001
42	Gatsibo_NLWH	n_season_fert	1.517	0.296	< 0.001
428	Rutsiro	own.cows	0.824	0.545	< 0.001
258	Nyamagabe	n_season_fert	4.000	1.554	< 0.001
122	Huye	valley	0.254	0.052	< 0.001
123	Huye	hillside	0.702	0.922	< 0.001
296	Nyamasheke	n_season_lime	0.327	0.041	< 0.001
186	Kayonza	n_season_fert	2.073	0.776	0.001
434	Rwamagana	age	44.245	52.072	0.002
291	Nyamasheke	hhsize	5.701	4.685	0.005
145	Karongi	female	0.466	0.667	0.007
363	Nyaruguru	hhsize	6.217	4.717	0.007
104	Gisagara	own.cows	0.696	0.326	0.007
106	Gisagara	own.chickens	0.370	0.065	0.009
107	Gisagara	own.pigs	0.565	0.217	0.011
110	Huye	age	43.395	49.400	0.015
89	Gisagara	cows	1.043	0.413	0.016
99	Gisagara	m3.Mg	191.739	239.872	0.02
325	Nyanza	female	0.217	0.543	0.021
98	Gisagara	m3.Ca	903.287	1203.933	0.021
330	Nyanza	n_season_fert	1.326	0.152	0.03
38	Gatsibo_NLWH	age	39.897	46.210	0.032
37	Gatsibo_NLWH	female	0.322	0.556	0.033
91	Gisagara	chickens	1.196	0.174	0.033
466	Rwamagana	own.chickens	0.418	0.225	0.033
2	Gatsibo_LWH	age	41.931	49.949	0.042
253	Nyamagabe	female	0.482	0.750	0.047
181	Kayonza	female	0.400	0.672	0.048
140	Huye	own.cows	0.675	0.487	0.05
40	Gatsibo_NLWH	own	0.908	1.000	0.051
224	Mugonero	n_season_lime	0.176	0.023	0.051
433	Rwamagana	female	0.555	0.739	0.051
1	Gatsibo_LWH	female	0.241	0.492	0.056
399	Rutsiro	hhsize	5.624	4.844	0.056
310	Nyamasheke	climbing_bean	0.422	0.267	0.06
284	Nyamagabe	own.cows	0.732	0.482	0.071
313	Nyamasheke	maize	0.082	0.014	0.071
169	Karongi	maize	0.086	0.019	0.072
219	Mugonero	hhsize	5.802	5.054	0.074
59	Gatsibo_NLWH	sorghum	0.276	0.469	0.098
218	Mugonero	age	44.290	49.140	0.1
101	Gisagara	Total.N	0.142	0.155	0.116
125	Huye	cows	1.167	0.774	0.116
361	Nyaruguru	female	0.391	0.652	0.116
300	Nyamasheke	slope	14.129	17.199	0.123
404	Rutsiro	n_season_lime	0.327	0.096	0.123
102	Gisagara	Total.C	1.815	2.005	0.124
295	Nyamasheke	n_season_compost	6.544	5.521	0.131
338	Nyanza	valley	0.283	0.087	0.131
339	Nyanza	hillside	0.717	0.913	0.131
298	Nyamasheke	n_seasons_leg_1	2.327	1.658	0.149
464	Rwamagana	own.cows	0.555	0.396	0.149
238	Mugonero	climbing_bean	0.321	0.194	0.152
366	Nyaruguru	n_season_fert	2.130	1.065	0.152
92	Gisagara	pigs	0.609	0.304	0.159
128	Huye	pigs	0.965	0.748	0.159
75	Gisagara	hhsize	5.348	4.326	0.162
6	Gatsibo_LWH	n_season_fert	2.552	1.475	0.164
119	Huye	n_seasons_leg_2	3.667	4.860	0.164
131	Huye	sorghum	0.237	0.374	0.164
336	Nyanza	slope	7.435	9.239	0.164
396	Nyaruguru	own.sheep	0.109	0.000	0.164
455	Rwamagana	sorghum	0.191	0.324	0.164
143	Huye	own.pigs	0.754	0.617	0.17
417	Rutsiro	sheep	0.467	0.246	0.177
111	Huye	hhsize	5.132	4.574	0.187
255	Nyamagabe	hhsize	5.482	4.696	0.187
309	Nyamasheke	sheep	0.082	0.014	0.197
271	Nyamagabe	chickens	0.857	0.304	0.202
430	Rutsiro	own.chickens	0.291	0.192	0.212
286	Nyamagabe	own.chickens	0.286	0.125	0.216
408	Rutsiro	slope	15.521	13.958	0.22
179	Karongi	own.pigs	0.399	0.289	0.225
192	Kayonza	slope	3.327	1.690	0.225
81	Gisagara	n_season_fallow	0.500	0.087	0.226
83	Gisagara	n_seasons_leg_2	2.370	3.283	0.226
227	Mugonero	n_seasons_leg_2	3.131	4.031	0.23
56	Gatsibo_NLWH	pigs	0.241	0.062	0.236
100	Gisagara	pH	5.759	5.950	0.236
328	Nyanza	own	0.913	1.000	0.238
436	Rwamagana	own	0.909	0.973	0.238
103	Gisagara	ExAl	0.337	0.191	0.254
412	Rutsiro	hilltop	0.097	0.042	0.254
449	Rwamagana	cows	1.082	0.514	0.254
467	Rwamagana	own.pigs	0.282	0.171	0.254
451	Rwamagana	chickens	1.609	0.955	0.26
47	Gatsibo_NLWH	n_seasons_leg_2	3.046	3.889	0.262
78	Gisagara	n_season_fert	1.087	0.348	0.262
324	Nyamasheke	own.sheep	0.054	0.014	0.269
335	Nyanza	n_seasons_leg_2	2.130	3.478	0.269
152	Karongi	n_season_lime	0.098	0.013	0.271
188	Kayonza	n_season_lime	0.527	0.707	0.273
343	Nyanza	chickens	2.348	1.174	0.273
441	Rwamagana	n_season_fallow	1.218	0.712	0.273
381	Nyaruguru	sheep	0.196	0.000	0.274
90	Gisagara	goats	1.522	0.935	0.275
254	Nyamagabe	age	43.768	48.875	0.275
160	Karongi	hilltop	0.067	0.025	0.315
86	Gisagara	valley	0.283	0.130	0.315
273	Nyamagabe	sheep	0.286	0.089	0.315
377	Nyaruguru	cows	1.196	0.891	0.315
389	Nyaruguru	Total.N	0.161	0.170	0.315
323	Nyamasheke	own.pigs	0.408	0.308	0.318
137	Huye	Total.N	0.142	0.148	0.327
161	Karongi	cows	1.387	1.119	0.348
355	Nyanza	ExAl	0.117	0.175	0.358
52	Gatsibo_NLWH	hilltop	0.011	0.062	0.358
133	Huye	maize	0.061	0.017	0.358
201	Kayonza	sheep	0.036	0.172	0.358
326	Nyanza	age	44.717	50.152	0.372
392	Nyaruguru	own.cows	0.826	0.674	0.372
73	Gisagara	female	0.435	0.609	0.377
395	Nyaruguru	own.pigs	0.609	0.435	0.377
146	Karongi	age	44.681	47.384	0.379
210	Kayonza	Total.C	2.395	2.552	0.382
380	Nyaruguru	pigs	0.783	0.457	0.382
459	Rwamagana	m3.Mg	270.948	289.488	0.384
248	Mugonero	own.cows	0.786	0.698	0.387
182	Kayonza	age	42.345	46.500	0.389
403	Rutsiro	n_season_compost	8.176	7.593	0.417
43	Gatsibo_NLWH	n_season_compost	2.632	3.420	0.417
33	Gatsibo_LWH	own.goats	0.552	0.407	0.429
206	Kayonza	m3.Ca	1677.807	1878.700	0.432
348	Nyanza	bush_bean	0.261	0.413	0.446
168	Karongi	bush_bean	0.080	0.132	0.454
413	Rutsiro	cows	1.442	1.048	0.458
394	Nyaruguru	own.chickens	0.283	0.152	0.46
242	Mugonero	m3.Ca	551.499	603.770	0.462
405	Rutsiro	n_season_fallow	0.473	0.287	0.485
299	Nyamasheke	n_seasons_leg_2	2.137	2.681	0.488
71	Gatsibo_NLWH	own.pigs	0.126	0.062	0.499
144	Huye	own.sheep	0.053	0.017	0.499
272	Nyamagabe	pigs	1.000	0.804	0.499
391	Nyaruguru	ExAl	0.717	0.882	0.499
32	Gatsibo_LWH	own.cows	0.500	0.373	0.5
96	Gisagara	bush_bean	0.217	0.348	0.5
97	Gisagara	maize	0.043	0.000	0.5
109	Huye	female	0.500	0.591	0.5
184	Kayonza	own	0.964	0.897	0.5
185	Kayonza	field.size	664.236	423.776	0.5
216	Kayonza	own.sheep	0.036	0.103	0.5
232	Mugonero	hilltop	0.015	0.000	0.5
262	Nyamagabe	n_seasons_leg_1	1.625	1.125	0.5
276	Nyamagabe	bush_bean	0.000	0.036	0.5
288	Nyamagabe	own.sheep	0.179	0.089	0.5
352	Nyanza	pH	6.104	5.995	0.5
368	Nyaruguru	n_season_lime	0.043	0.000	0.5
388	Nyaruguru	pH	5.320	5.205	0.5
397	Rutsiro	female	0.545	0.623	0.5
452	Rwamagana	pigs	0.518	0.315	0.5
44	Gatsibo_NLWH	n_season_lime	0.069	0.025	0.51
17	Gatsibo_LWH	cows	0.759	0.542	0.514
49	Gatsibo_NLWH	alt	1017.040	1129.172	0.514
55	Gatsibo_NLWH	chickens	1.678	1.000	0.514
162	Karongi	goats	0.890	1.069	0.514
209	Kayonza	Total.N	0.180	0.187	0.514
344	Nyanza	pigs	0.087	0.217	0.514
359	Nyanza	own.pigs	0.065	0.152	0.514
364	Nyaruguru	own	0.848	0.935	0.514
462	Rwamagana	Total.C	2.174	2.236	0.514
205	Kayonza	maize	0.018	0.069	0.514
432	Rutsiro	own.sheep	0.212	0.156	0.514
424	Rutsiro	pH	5.269	5.339	0.518
87	Gisagara	hillside	0.587	0.717	0.524
151	Karongi	n_season_compost	7.074	6.553	0.524
199	Kayonza	chickens	0.982	0.466	0.544
422	Rutsiro	m3.Ca	628.725	682.389	0.549
290	Nyamasheke	age	45.898	48.123	0.551
356	Nyanza	own.cows	0.652	0.522	0.551
105	Gisagara	own.goats	0.630	0.500	0.552
148	Karongi	own	0.957	0.981	0.552
207	Kayonza	m3.Mg	331.923	356.976	0.552
244	Mugonero	pH	5.180	5.243	0.552
247	Mugonero	ExAl	0.925	0.823	0.552
320	Nyamasheke	own.cows	0.639	0.568	0.552
350	Nyanza	m3.Ca	1086.559	992.934	0.552
11	Gatsibo_LWH	n_seasons_leg_2	1.983	2.690	0.556
18	Gatsibo_LWH	goats	1.345	0.983	0.562
118	Huye	n_seasons_leg_1	1.386	1.043	0.562
194	Kayonza	valley	0.545	0.431	0.562
351	Nyanza	m3.Mg	233.315	213.344	0.562
19	Gatsibo_LWH	chickens	1.052	0.678	0.563
127	Huye	chickens	1.772	1.348	0.563
409	Rutsiro	alt	1945.122	1894.425	0.563
443	Rwamagana	n_seasons_leg_2	1.349	1.685	0.563
147	Karongi	hhsize	5.172	4.912	0.568
159	Karongi	hillside	0.785	0.836	0.568
203	Kayonza	sorghum	0.182	0.103	0.568
387	Nyaruguru	m3.Mg	152.908	139.740	0.568
390	Nyaruguru	Total.C	2.145	2.237	0.568
414	Rutsiro	goats	0.515	0.659	0.568
277	Nyamagabe	maize	0.089	0.036	0.569
426	Rutsiro	Total.C	2.247	2.309	0.569
444	Rwamagana	slope	3.682	4.234	0.569
429	Rutsiro	own.goats	0.273	0.329	0.602
316	Nyamasheke	pH	5.176	5.120	0.605
204	Kayonza	bush_bean	0.218	0.138	0.614
61	Gatsibo_NLWH	maize	0.092	0.049	0.632
250	Mugonero	own.chickens	0.382	0.318	0.632
427	Rutsiro	ExAl	0.814	0.736	0.632
453	Rwamagana	sheep	0.509	0.360	0.636
3	Gatsibo_LWH	hhsize	5.362	5.763	0.641
4	Gatsibo_LWH	own	0.983	1.000	0.641
54	Gatsibo_NLWH	goats	1.218	1.654	0.641
68	Gatsibo_NLWH	own.cows	0.563	0.481	0.641
82	Gisagara	n_seasons_leg_1	2.304	2.870	0.641
112	Huye	own	0.974	0.991	0.641
116	Huye	n_season_lime	0.035	0.113	0.641
126	Huye	goats	0.965	0.809	0.641
174	Karongi	Total.C	1.887	1.839	0.641
198	Kayonza	goats	1.727	1.397	0.641
202	Kayonza	climbing_bean	0.000	0.017	0.641
220	Mugonero	own	0.947	0.915	0.641
241	Mugonero	maize	0.000	0.008	0.641
269	Nyamagabe	cows	0.911	0.714	0.641
319	Nyamasheke	ExAl	0.983	1.062	0.641
349	Nyanza	maize	0.130	0.065	0.641
358	Nyanza	own.chickens	0.543	0.435	0.641
365	Nyaruguru	field.size	570.022	495.239	0.641
384	Nyaruguru	bush_bean	0.065	0.022	0.641
385	Nyaruguru	maize	0.022	0.000	0.641
401	Rutsiro	field.size	584.327	430.156	0.641
416	Rutsiro	pigs	0.315	0.240	0.641
457	Rwamagana	maize	0.027	0.009	0.641
458	Rwamagana	m3.Ca	1303.126	1380.293	0.641
461	Rwamagana	Total.N	0.175	0.179	0.641
217	Mugonero	female	0.656	0.713	0.644
341	Nyanza	cows	1.174	0.891	0.644
50	Gatsibo_NLWH	valley	0.759	0.691	0.646
166	Karongi	climbing_bean	0.405	0.352	0.646
233	Mugonero	cows	1.328	1.171	0.646
347	Nyanza	sorghum	0.283	0.196	0.646
410	Rutsiro	valley	0.152	0.192	0.646
196	Kayonza	hilltop	0.109	0.172	0.646
53	Gatsibo_NLWH	cows	0.885	0.741	0.651
301	Nyamasheke	alt	1678.150	1704.388	0.651
423	Rutsiro	m3.Mg	165.214	175.386	0.651
213	Kayonza	own.goats	0.673	0.586	0.652
8	Gatsibo_LWH	n_season_lime	0.500	0.322	0.653
77	Gisagara	field.size	493.957	559.261	0.661
132	Huye	bush_bean	0.193	0.243	0.661
281	Nyamagabe	Total.N	0.169	0.165	0.661
379	Nyaruguru	chickens	0.652	0.391	0.661
70	Gatsibo_NLWH	own.chickens	0.310	0.247	0.665
421	Rutsiro	maize	0.176	0.216	0.665
5	Gatsibo_LWH	field.size	776.345	916.441	0.688
85	Gisagara	alt	1036.055	923.525	0.688
282	Nyamagabe	Total.C	2.345	2.274	0.688
337	Nyanza	alt	1505.653	1529.554	0.688
208	Kayonza	pH	6.132	6.204	0.689
221	Mugonero	field.size	441.359	393.124	0.689
425	Rutsiro	Total.N	0.157	0.159	0.689
468	Rwamagana	own.sheep	0.227	0.180	0.689
34	Gatsibo_LWH	own.chickens	0.345	0.271	0.694
141	Huye	own.goats	0.518	0.461	0.694
45	Gatsibo_NLWH	n_season_fallow	0.736	0.543	0.695
308	Nyamasheke	pigs	0.776	0.582	0.695
393	Nyaruguru	own.goats	0.348	0.435	0.695
357	Nyanza	own.goats	0.630	0.543	0.697
375	Nyaruguru	hillside	0.957	0.913	0.697
376	Nyaruguru	hilltop	0.043	0.087	0.697
95	Gisagara	sorghum	0.435	0.522	0.699
270	Nyamagabe	goats	0.750	0.607	0.699
289	Nyamasheke	female	0.646	0.692	0.699
155	Karongi	n_seasons_leg_2	3.216	3.548	0.701
287	Nyamagabe	own.pigs	0.732	0.661	0.704
60	Gatsibo_NLWH	bush_bean	0.172	0.222	0.704
117	Huye	n_season_fallow	0.675	0.487	0.704
293	Nyamasheke	field.size	787.527	580.130	0.704
327	Nyanza	hhsize	5.261	4.891	0.704
386	Nyaruguru	m3.Ca	660.083	610.274	0.704
236	Mugonero	pigs	0.412	0.333	0.709
212	Kayonza	own.cows	0.382	0.310	0.709
177	Karongi	own.goats	0.497	0.541	0.71
251	Mugonero	own.pigs	0.244	0.287	0.718
373	Nyaruguru	alt	1791.131	1814.738	0.718
257	Nyamagabe	field.size	310.929	282.643	0.721
74	Gisagara	age	44.761	46.848	0.726
252	Mugonero	own.sheep	0.122	0.093	0.726
456	Rwamagana	bush_bean	0.364	0.315	0.726
243	Mugonero	m3.Mg	160.377	170.165	0.732
10	Gatsibo_LWH	n_seasons_leg_1	4.345	4.678	0.741
23	Gatsibo_LWH	sorghum	0.276	0.339	0.741
26	Gatsibo_LWH	m3.Ca	1226.445	1149.970	0.741
124	Huye	hilltop	0.044	0.026	0.741
149	Karongi	field.size	490.788	446.088	0.741
164	Karongi	pigs	0.528	0.447	0.741
175	Karongi	ExAl	0.647	0.597	0.745
200	Kayonza	pigs	0.309	0.224	0.746
129	Huye	sheep	0.061	0.035	0.747
460	Rwamagana	pH	5.817	5.862	0.756
41	Gatsibo_NLWH	field.size	1306.069	1459.901	0.764
115	Huye	n_season_compost	5.211	4.878	0.764
24	Gatsibo_LWH	bush_bean	0.397	0.339	0.779
57	Gatsibo_NLWH	sheep	0.046	0.025	0.779
69	Gatsibo_NLWH	own.goats	0.517	0.568	0.779
171	Karongi	m3.Mg	254.160	267.784	0.779
173	Karongi	Total.N	0.144	0.142	0.779
176	Karongi	own.cows	0.767	0.736	0.779
190	Kayonza	n_seasons_leg_1	1.982	2.155	0.779
317	Nyamasheke	Total.N	0.165	0.167	0.779
333	Nyanza	n_season_fallow	1.130	0.870	0.779
400	Rutsiro	own	0.915	0.934	0.779
20	Gatsibo_LWH	pigs	0.138	0.203	0.782
79	Gisagara	n_season_compost	4.326	3.957	0.782
214	Kayonza	own.chickens	0.182	0.138	0.782
172	Karongi	pH	5.442	5.477	0.785
439	Rwamagana	n_season_compost	4.018	3.748	0.785
36	Gatsibo_LWH	own.sheep	0.034	0.017	0.792
39	Gatsibo_NLWH	hhsize	5.333	5.160	0.792
191	Kayonza	n_seasons_leg_2	1.091	1.259	0.792
215	Kayonza	own.pigs	0.218	0.172	0.792
226	Mugonero	n_seasons_leg_1	1.336	1.186	0.792
256	Nyamagabe	own	0.964	0.982	0.792
345	Nyanza	sheep	0.022	0.043	0.792
346	Nyanza	climbing_bean	0.043	0.022	0.792
360	Nyanza	own.sheep	0.022	0.043	0.792
369	Nyaruguru	n_season_fallow	0.370	0.261	0.792
378	Nyaruguru	goats	0.609	0.739	0.792
446	Rwamagana	valley	0.409	0.369	0.792
447	Rwamagana	hillside	0.527	0.568	0.792
450	Rwamagana	goats	1.745	1.568	0.792
76	Gisagara	own	0.826	0.870	0.793
9	Gatsibo_LWH	n_season_fallow	0.397	0.288	0.796
259	Nyamagabe	n_season_compost	7.857	7.518	0.796
285	Nyamagabe	own.goats	0.464	0.411	0.796
398	Rutsiro	age	43.685	44.617	0.796
195	Kayonza	hillside	0.345	0.397	0.797
84	Gisagara	slope	8.391	7.739	0.802
315	Nyamasheke	m3.Mg	156.068	149.714	0.802
260	Nyamagabe	n_season_lime	0.500	0.393	0.803
283	Nyamagabe	ExAl	1.111	1.170	0.808
134	Huye	m3.Ca	850.319	872.728	0.815
135	Huye	m3.Mg	187.671	191.878	0.815
63	Gatsibo_NLWH	m3.Mg	246.996	255.132	0.818
249	Mugonero	own.goats	0.542	0.574	0.818
265	Nyamagabe	alt	2124.389	2109.814	0.818
113	Huye	field.size	599.096	567.922	0.821
263	Nyamagabe	n_seasons_leg_2	3.161	2.929	0.821
278	Nyamagabe	m3.Ca	458.692	440.797	0.821
382	Nyaruguru	climbing_bean	0.196	0.239	0.821
314	Nyamasheke	m3.Ca	614.196	590.190	0.84
180	Karongi	own.sheep	0.080	0.094	0.84
228	Mugonero	slope	15.939	16.473	0.84
292	Nyamasheke	own	0.932	0.945	0.84
367	Nyaruguru	n_season_compost	5.696	6.043	0.84
465	Rwamagana	own.goats	0.536	0.568	0.84
142	Huye	own.chickens	0.421	0.391	0.843
275	Nyamagabe	sorghum	0.036	0.054	0.843
329	Nyanza	field.size	853.359	766.217	0.843
235	Mugonero	chickens	1.076	0.961	0.844
370	Nyaruguru	n_seasons_leg_1	2.043	2.348	0.844
334	Nyanza	n_seasons_leg_1	2.543	2.283	0.856
178	Karongi	own.chickens	0.411	0.434	0.863
437	Rwamagana	field.size	892.855	842.054	0.863
442	Rwamagana	n_seasons_leg_1	1.682	1.568	0.863
312	Nyamasheke	bush_bean	0.088	0.075	0.867
163	Karongi	chickens	1.503	1.377	0.87
440	Rwamagana	n_season_lime	0.355	0.324	0.87
454	Rwamagana	climbing_bean	0.036	0.027	0.872
72	Gatsibo_NLWH	own.sheep	0.034	0.025	0.89
306	Nyamasheke	goats	0.918	0.863	0.89
165	Karongi	sheep	0.166	0.195	0.902
13	Gatsibo_LWH	alt	1094.532	1126.650	0.912
136	Huye	pH	5.681	5.664	0.912
223	Mugonero	n_season_compost	6.229	6.070	0.912
297	Nyamasheke	n_season_fallow	0.599	0.664	0.912
431	Rutsiro	own.pigs	0.194	0.180	0.912
35	Gatsibo_LWH	own.pigs	0.121	0.102	0.916
411	Rutsiro	hillside	0.752	0.766	0.919
231	Mugonero	hillside	0.855	0.868	0.925
31	Gatsibo_LWH	ExAl	0.151	0.164	0.927
28	Gatsibo_LWH	pH	6.139	6.115	0.928
29	Gatsibo_LWH	Total.N	0.148	0.147	0.928
30	Gatsibo_LWH	Total.C	1.882	1.856	0.928
88	Gisagara	hilltop	0.130	0.152	0.928
120	Huye	slope	6.658	6.817	0.928
266	Nyamagabe	valley	0.125	0.107	0.928
121	Huye	alt	1647.892	1655.403	0.93
318	Nyamasheke	Total.C	2.307	2.289	0.93
371	Nyaruguru	n_seasons_leg_2	3.178	3.370	0.93
445	Rwamagana	alt	982.728	961.878	0.93
51	Gatsibo_NLWH	hillside	0.230	0.247	0.934
64	Gatsibo_NLWH	pH	6.037	6.053	0.934
170	Karongi	m3.Ca	781.863	796.295	0.934
274	Nyamagabe	climbing_bean	0.161	0.143	0.934
280	Nyamagabe	pH	4.999	5.013	0.934
267	Nyamagabe	hillside	0.821	0.839	0.938
406	Rutsiro	n_seasons_leg_1	3.388	3.479	0.938
158	Karongi	valley	0.147	0.138	0.944
197	Kayonza	cows	0.727	0.828	0.944
225	Mugonero	n_season_fallow	0.863	0.915	0.944
239	Mugonero	sorghum	0.061	0.054	0.944
353	Nyanza	Total.N	0.134	0.135	0.944
27	Gatsibo_LWH	m3.Mg	238.242	234.655	0.946
130	Huye	climbing_bean	0.088	0.096	0.946
211	Kayonza	ExAl	0.111	0.117	0.946
264	Nyamagabe	slope	12.571	12.839	0.946
279	Nyamagabe	m3.Mg	109.343	111.050	0.946
307	Nyamasheke	chickens	0.599	0.562	0.946
463	Rwamagana	ExAl	0.257	0.249	0.946
67	Gatsibo_NLWH	ExAl	0.182	0.175	0.947
261	Nyamagabe	n_season_fallow	0.268	0.304	0.947
354	Nyanza	Total.C	1.766	1.782	0.947
139	Huye	ExAl	0.307	0.315	0.954
435	Rwamagana	hhsize	4.927	4.982	0.954
415	Rutsiro	chickens	0.879	0.958	0.955
138	Huye	Total.C	1.917	1.925	0.956
62	Gatsibo_NLWH	m3.Ca	1236.151	1251.746	0.956
183	Kayonza	hhsize	5.091	5.155	0.956
21	Gatsibo_LWH	sheep	0.103	0.085	0.96
418	Rutsiro	climbing_bean	0.327	0.335	0.965
7	Gatsibo_LWH	n_season_compost	5.672	5.780	0.97
229	Mugonero	alt	1796.175	1788.315	0.97
48	Gatsibo_NLWH	slope	4.011	4.099	0.971
66	Gatsibo_NLWH	Total.C	2.002	1.990	0.971
189	Kayonza	n_season_fallow	0.455	0.483	0.971
302	Nyamasheke	valley	0.150	0.144	0.971
407	Rutsiro	n_seasons_leg_2	2.036	2.078	0.971
303	Nyamasheke	hillside	0.823	0.829	0.971
14	Gatsibo_LWH	valley	0.603	0.593	0.979
15	Gatsibo_LWH	hillside	0.397	0.407	0.979
237	Mugonero	sheep	0.221	0.233	0.984
65	Gatsibo_NLWH	Total.N	0.157	0.158	0.985
167	Karongi	sorghum	0.147	0.151	0.986
234	Mugonero	goats	1.321	1.302	0.986
331	Nyanza	n_season_compost	3.261	3.326	0.986
342	Nyanza	goats	1.261	1.283	0.997
245	Mugonero	Total.N	0.153	0.153	0.999
12	Gatsibo_LWH	slope	4.259	4.220	1
46	Gatsibo_NLWH	n_seasons_leg_1	1.540	1.556	1
80	Gisagara	n_season_lime	0.022	0.022	1
93	Gisagara	sheep	0.022	0.022	1
94	Gisagara	climbing_bean	0.022	0.022	1
108	Gisagara	own.sheep	0.022	0.022	1
153	Karongi	n_season_fallow	0.834	0.843	1
154	Karongi	n_seasons_leg_1	2.485	2.497	1
156	Karongi	slope	11.975	11.937	1
157	Karongi	alt	1806.841	1806.879	1
187	Kayonza	n_season_compost	3.564	3.534	1
193	Kayonza	alt	1196.807	1202.646	1
230	Mugonero	valley	0.130	0.132	1
240	Mugonero	bush_bean	0.176	0.178	1
246	Mugonero	Total.C	2.217	2.218	1
268	Nyamagabe	hilltop	0.054	0.054	1
304	Nyamasheke	hilltop	0.027	0.027	1
305	Nyamasheke	cows	1.245	1.240	1
321	Nyamasheke	own.goats	0.442	0.445	1
322	Nyamasheke	own.chickens	0.218	0.219	1
362	Nyaruguru	age	48.457	48.304	1
372	Nyaruguru	slope	9.478	9.457	1
383	Nyaruguru	sorghum	0.304	0.304	1
420	Rutsiro	bush_bean	0.012	0.012	1
448	Rwamagana	hilltop	0.064	0.063	1
16	Gatsibo_LWH	hilltop	0.000	0.000	NA
22	Gatsibo_LWH	climbing_bean	0.000	0.000	NA
25	Gatsibo_LWH	maize	0.000	0.000	NA
58	Gatsibo_NLWH	climbing_bean	0.000	0.000	NA
311	Nyamasheke	sorghum	0.000	0.000	NA
332	Nyanza	n_season_lime	0.000	0.000	NA
340	Nyanza	hilltop	0.000	0.000	NA
374	Nyaruguru	valley	0.000	0.000	NA
419	Rutsiro	sorghum	0.000	0.000	NA

District balance interpretation

Demographic variables interpretation here.

Agricultural practice variables interpretation here

Soil Variables interpretation here

write.csv(dist.output, file=paste(od, "district balance.csv", sep="/"), row.names=T)

Baseline balance by Tubura tenure

Look at farmers by duration of tenure farming with Tubura. We want to understand, at least with an initial naive baseline sense, what is the cumulative effect of Tubura practices on soil health outcomes?

We will look only at current Tubura farmers and compare first year farmers to farmers with more experience with Tubura.

oafOnly <- d[which(d$client==1 & d$total.seasons>=1),]

nTenure <- oafOnly %>% group_by(total.seasons) %>% 
  dplyr::summarize(
    n = n()
  ) %>% ungroup() %>% as.data.frame()
nTenure$val <- paste(nTenure$total.seasons, " (", "n = ", nTenure$n, ")", sep = "")

for(i in 1:length(soilVars)){
print(
  ggplot(oafOnly, aes(x=as.factor(total.seasons), y=oafOnly[,soilVars[i]])) + 
    geom_boxplot() +
    scale_x_discrete(labels=nTenure$val) + 
    theme(legend.position = "bottom", axis.text.x = element_text(angle = 45, hjust = 1)) +
    labs(x="Tubura Tenure", y=soilVars[i], title = paste("RW baseline soil by tenure - ", soilVars[i], sep = ""))
  )  
}

Tenure summaries

tenureSum <- aggregate(oafOnly[,out.list], by=list(oafOnly$total.seasons), function(x){
  round(mean(x, na.rm=T),2)
})
tenureSum <- as.data.frame(t(tenureSum))
colnames(tenureSum) <- c(paste(seq(1,14,1), " seas.", sep = ""))

print(kable(tenureSum))

	1 seas.	2 seas.	3 seas.	4 seas.	5 seas.	6 seas.	7 seas.	8 seas.	9 seas.	10 seas.	11 seas.	12 seas.	13 seas.	14 seas.
Group.1	1.00	2.00	3.00	4.00	5.00	6.00	7.00	8.00	9.00	10.00	11.00	12.00	13.00	14.00
female	0.42	0.48	0.60	0.49	0.59	0.59	0.52	0.50	0.65	0.62	0.75	0.53	0.55	0.48
age	42.63	42.68	44.33	45.58	43.28	43.96	44.19	50.91	51.85	42.76	48.25	48.97	52.73	49.52
hhsize	5.24	5.17	5.08	5.21	6.07	5.58	6.19	5.66	6.65	6.07	6.58	5.53	4.55	6.12
own	0.93	0.95	0.92	0.95	0.98	0.97	1.00	0.93	0.85	1.00	0.92	0.89	0.91	0.95
field.size	606.74	630.98	630.65	551.74	482.02	484.79	412.52	592.95	330.00	614.79	617.83	440.76	280.82	679.71
n_season_fert	1.76	2.32	2.63	3.28	4.67	4.90	6.07	5.59	6.00	5.83	7.25	7.18	7.18	6.79
n_season_compost	5.35	5.37	4.81	6.37	6.87	7.13	8.30	7.43	7.30	7.52	8.67	8.24	8.27	7.81
n_season_lime	0.16	0.27	0.36	0.31	0.22	0.26	0.07	0.07	0.20	0.28	0.08	0.16	0.09	0.48
n_season_fallow	0.65	0.50	0.77	0.90	0.65	0.66	0.67	0.77	1.10	0.21	0.42	0.21	1.82	0.67
n_seasons_leg_1	1.97	2.07	2.10	2.44	2.96	2.89	1.96	2.91	1.60	1.79	2.92	3.53	2.09	3.02
n_seasons_leg_2	2.44	2.54	2.03	2.32	2.67	2.62	4.93	3.50	2.75	2.59	3.92	1.49	2.82	2.12
slope	9.05	8.66	9.65	8.80	9.57	11.84	14.07	11.77	13.60	12.31	14.17	11.26	14.91	14.38
alt	1531.01	1491.56	1550.41	1518.25	1654.93	1582.27	1758.60	1639.74	1812.98	1764.71	1788.95	1640.20	1702.35	1633.76
valley	0.31	0.32	0.27	0.25	0.30	0.22	0.26	0.23	0.00	0.07	0.33	0.18	0.09	0.21
hillside	0.62	0.61	0.71	0.72	0.67	0.74	0.70	0.66	0.95	0.90	0.67	0.79	0.82	0.76
hilltop	0.07	0.06	0.03	0.03	0.02	0.05	0.04	0.11	0.05	0.03	0.00	0.03	0.09	0.02
cows	0.99	1.00	0.94	1.49	1.91	1.28	1.30	1.09	3.25	1.31	1.75	0.97	1.00	1.07
goats	0.97	1.12	1.00	1.28	1.39	1.00	0.78	1.16	1.20	0.90	0.83	1.11	0.45	1.26
chickens	1.07	1.22	1.08	1.20	1.28	0.87	1.78	1.61	1.20	1.45	3.58	1.76	0.45	1.00
pigs	0.55	0.52	0.58	0.51	0.37	0.47	0.48	0.55	0.45	0.55	0.25	0.84	0.64	0.83
sheep	0.20	0.21	0.29	0.34	0.17	0.16	0.44	0.23	0.15	0.41	0.00	0.05	0.18	0.17
climbing_bean	0.14	0.17	0.21	0.17	0.20	0.32	0.52	0.36	0.35	0.45	0.33	0.39	0.55	0.38
sorghum	0.18	0.17	0.12	0.14	0.11	0.06	0.04	0.02	0.05	0.10	0.17	0.05	0.09	0.00
bush_bean	0.14	0.17	0.17	0.17	0.15	0.19	0.04	0.18	0.15	0.03	0.08	0.11	0.18	0.14
maize	0.06	0.07	0.06	0.08	0.09	0.08	0.11	0.11	0.10	0.10	0.08	0.03	0.00	0.05
m3.Ca	890.32	911.95	871.92	918.19	913.35	840.76	771.94	755.95	592.43	582.74	663.89	848.28	592.16	618.11
m3.Mg	195.33	201.53	213.24	210.61	218.70	214.24	210.69	222.18	156.45	172.28	198.05	275.01	151.55	166.45
pH	5.57	5.57	5.57	5.53	5.64	5.52	5.43	5.46	5.19	5.22	5.37	5.57	5.29	5.25
Total.N	0.16	0.16	0.16	0.16	0.15	0.15	0.15	0.14	0.16	0.16	0.14	0.15	0.15	0.16
Total.C	2.11	2.15	2.11	2.15	2.13	2.07	2.05	1.93	2.38	2.16	1.88	1.97	2.00	2.22
ExAl	0.59	0.57	0.62	0.54	0.46	0.61	0.60	0.56	0.91	0.75	0.70	0.48	0.64	0.90
own.cows	0.69	0.63	0.58	0.75	0.72	0.69	0.81	0.70	0.90	0.90	0.75	0.63	0.64	0.81
own.goats	0.45	0.52	0.47	0.46	0.54	0.50	0.48	0.55	0.50	0.48	0.50	0.53	0.27	0.52
own.chickens	0.30	0.35	0.33	0.30	0.33	0.30	0.56	0.43	0.45	0.45	0.58	0.42	0.45	0.33
own.pigs	0.42	0.36	0.32	0.31	0.26	0.33	0.41	0.36	0.35	0.34	0.17	0.39	0.36	0.40
own.sheep	0.13	0.09	0.14	0.17	0.11	0.08	0.26	0.14	0.10	0.14	0.00	0.03	0.09	0.07

Tenure balance

We’re defining Tubura tenure as having 3 or more seasons of experience farming with Tubura. We draw the line at 3 seasons as three seasons of fertilizer use is approximately when we’d expect fertilizer to start to have a detrimental effect on soil health.

oafOnly$tenured <- ifelse(oafOnly$total.seasons>=3,1,0)

tenure <- do.call(rbind, lapply(out.list, function(x) {
  
  out <- t.test(oafOnly[,x] ~ oafOnly[,"tenured"], data=oafOnly)
  tab <- data.frame(out[[5]][[1]], out[[5]][[2]], out[3])
  tab[,1:2] <- round(tab[,1:2],3)
  names(tab) <- c(names(out[[5]]), "pvalue")
  return(tab)
}))

# use p.adjust with bonferroni correction
tenure$pvalue <- p.adjust(tenure$pvalue, method="fdr")
rownames(tenure) <- out.list
tenure <- tenure[order(tenure$pvalue),]
tenure$pvalue <- ifelse(tenure[, 3] < 0.001, "< 0.001", round(tenure[, 3], 3)) 


colnames(tenure) <- c("Non-Tubura", "Tubura Client", "p-value") 

print(kable(tenure))

	Non-Tubura	Tubura Client	p-value
n_season_fert	2.097	4.774	< 0.001
n_season_compost	5.359	6.899	< 0.001
climbing_bean	0.159	0.295	< 0.001
age	42.657	46.057	< 0.001
slope	8.814	11.103	< 0.001
hillside	0.616	0.736	< 0.001
sorghum	0.173	0.086	< 0.001
n_seasons_leg_1	2.030	2.583	0.006
valley	0.319	0.225	0.007
hhsize	5.200	5.597	0.01
cows	0.992	1.358	0.01
female	0.457	0.552	0.012
alt	1507.236	1611.057	0.015
m3.Ca	903.359	813.631	0.034
pH	5.570	5.478	0.036
Total.N	0.159	0.154	0.043
own.cows	0.657	0.719	0.097
field.size	621.349	535.512	0.117
hilltop	0.065	0.039	0.178
own.pigs	0.386	0.334	0.188
n_season_fallow	0.559	0.719	0.209
m3.Mg	199.065	208.991	0.334
Total.C	2.136	2.102	0.486
ExAl	0.574	0.613	0.518
own.sheep	0.103	0.122	0.518
own.chickens	0.332	0.360	0.525
sheep	0.203	0.242	0.598
maize	0.065	0.077	0.6
n_season_lime	0.227	0.259	0.621
chickens	1.162	1.261	0.633
goats	1.062	1.111	0.726
own	0.941	0.947	0.767
n_seasons_leg_2	2.500	2.575	0.767
bush_bean	0.159	0.151	0.767
pigs	0.532	0.545	0.871
own.goats	0.492	0.492	0.991

Tenured v. new balance interpretation

Demographic variables We are well balanced along demographic variables.

Agricultural practice variables Not surprisingly, Tubura farmers have more cumulative years of fertilizer use than current non-Tubura farmers. While that difference is signficant, it is realistically only a single season of fertilizer use different.

Interestingly, non-Tubura farmers reported using more lime than current Tubura farmers. This

Soil Variables Soil pH, calcium and magnesium levels are lower for tenured Tubura farmers. This is consistent with the hypothesis that increaesd fertilizer use leads to an increaese in soil acidity.

Analysis of agronomic practices contributing to soil health outcomes

Here’s where we’ll look at the contribution of fertilizer, lime and cultivation practices on soil health outcomes. This analysis will be come richer as we gain longitudinal measures. I caution that we cannot treat these relationships as causal. The direction of causality is not clearly delineated in the data or the study design. However, we can identify meaningful connections between practices and outcomes through this analysis to generate new hypotheses for field testing.

I’m going to start with behaviors by sections and then move to a more comprehensive model including multiple practices. All models will include controls for site to account for local variation and field officer behavior.

Agronomic Behaviors

Check for multicollinearity before adding number of seasons of agronomic inputs on the same side of the regression.

suppressMessages(library(stargazer))

inputUse <- c("n_season_fert","n_season_compost", "n_season_lime", "n_season_fallow")

cor(d[,inputUse], use="complete.obs")

##                  n_season_fert n_season_compost n_season_lime
## n_season_fert       1.00000000       0.35646933    0.17075604
## n_season_compost    0.35646933       1.00000000    0.03828723
## n_season_lime       0.17075604       0.03828723    1.00000000
## n_season_fallow    -0.06439632      -0.16923273   -0.01116383
##                  n_season_fallow
## n_season_fert        -0.06439632
## n_season_compost     -0.16923273
## n_season_lime        -0.01116383
## n_season_fallow       1.00000000

Interpretation: The strongest correlation between the input use intensity variables is between seasons of fertilizer and compost use, ~0.35. While this is on the higher end it’s not necessarily cause for alarm.

inputUse <- paste(c("n_season_fert","n_season_compost", "n_season_lime", "n_season_fallow"), collapse= " + ")

list1 <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~", inputUse, "+ as.factor(cell)", sep="")), data=d)
  return(mod)
})

stargazer(list1, type="html", 
          title = "2016A Rwanda Soil Health Baseline - Naive Agronomic Practice Models",
          covariate.labels = c("Seasons of Fertilizer", "Seasons of Compost", 
                               "Seasons of Lime", "Seasons of Fallow"),
          dep.var.caption = "",
          dep.var.labels = "",
          column.labels = c(gsub("m3.","", soilVars)),
          notes = "Includes FE for cell",
          omit=c("cell"), out=paste(od, "rw_baseline_agprac.htm", sep="/"))

2016A Rwanda Soil Health Baseline - Naive Agronomic Practice Models

	Ca	Mg	pH	Total.N	Total.C	ExAl
	(1)	(2)	(3)	(4)	(5)	(6)
Seasons of Fertilizer	-5.326	-1.128	-0.001	-0.0004**	-0.003	-0.001
	(3.321)	(0.702)	(0.003)	(0.0002)	(0.003)	(0.004)
Seasons of Compost	1.727	0.363	0.004	0.0001	0.002	-0.006**
	(2.680)	(0.567)	(0.003)	(0.0001)	(0.003)	(0.003)
Seasons of Lime	41.659***	7.998***	0.024*	0.002**	0.022	0.009
	(14.473)	(3.061)	(0.014)	(0.001)	(0.015)	(0.016)
Seasons of Fallow	-19.069***	-3.517***	-0.028***	0.0005	0.009*	0.025***
	(5.283)	(1.117)	(0.005)	(0.0003)	(0.005)	(0.006)
Constant	548.378	68.041	4.684***	0.199***	3.360***	1.904***
	(387.623)	(81.977)	(0.370)	(0.021)	(0.392)	(0.420)
Observations	2,443	2,443	2,443	2,443	2,443	2,443
R2	0.555	0.585	0.612	0.502	0.456	0.570
Adjusted R2	0.514	0.547	0.577	0.456	0.407	0.531
Residual Std. Error (df = 2238)	387.475	81.946	0.370	0.021	0.392	0.420
F Statistic (df = 204; 2238)	13.684***	15.452***	17.322***	11.044***	9.202***	14.545***
Note:	p<0.1; p<0.05; p<0.01
	Includes FE for cell

Interpretation The naive model suggests that when we include site level fixed effects, duration of agronomic practices don’t have a big effect on soil health outcomes. However, some of the practice intensity variables are not well distributed. Let’s take a look at a log transformation. I’m adding one to the variables as to not end up with lots of Inf values.

Log transformations in theory are appropriate for variables that are right skewed (vavlues clustered to the left of the distribution) and see diminishing returns to increasing values. The shape of the data suggests a log transformation but it’s debateable whether the relationship is diminishing.

agPrac <- c(names(d[grep('n_season_', names(d))]))

for(i in 1:length(agPrac)){
  print(
    ggplot(d, aes(x=d[,agPrac[i]])) + geom_density() +
      labs(x = paste(agPrac[i], " No transform", sep = ""))
        )
}

# since these are all skewed, consider a log transform 
for(i in 1:length(agPrac)){
  print(
    ggplot(d, aes(x=log10(d[,agPrac[i]]+1))) + geom_density() + 
      labs(x = paste(agPrac[i], " Log transform", sep = ""))
        )
}

# look at other transfomations
for(i in 1:length(agPrac)){
  print(
    ggplot(d, aes(x=d[,agPrac[i]]^(1/3))) + geom_density() +
      labs(x = paste(agPrac[i], " cubic root transform", sep = ""))
        )
}

# visualize the outcomes as well to see if a transformation is warranted
for(i in 1:length(soilVars)){
  print(
    ggplot(d, aes(x=d[,soilVars[i]])) + geom_density() +
      labs(x = soilVars[i], title = soilVars[i])
        )
}

d$logFert <- log(d$n_season_fert+1)
d$logCompost <- log(d$n_season_compost+1)
d$logLime <- log(d$n_season_lime+1)
d$logFallow <- log(d$n_season_fallow+1)

Or look at BoxCox graph to empirically determine the right transformation. Log is assuming a diminishing return to an increasing X. That’s probably not the case with fertilizer. We’d actually expect an increasing return as values get larger. We use boxcox to see what the data suggest. We interpret it as follows:

• lambda = 2 -> square
• lambda = 1 -> no transformation
• lambda = 0.5 -> square root
• lambda = 0 -> log
• lamdba = -1 -> inverse

library(MASS)
for(i in 1:length(agPrac)){
boxcox(lm(pH ~ d[,agPrac[i]], data=d))
}

For pH at least it seems like a log transform is appropriate. We can run this for all other variables as well to see what we get as well.

Let’s look at the log results: See here and here for guidance on intepreting log transformed right hand side variables. See here for additional guidance on choosing a transformation.

How to interpret RHS log transform: For a linear multivariate OLS regression, we say “a one unit increase in X causes a (coefficient) change in Y.” For a linear-log regression where the X variable is log transformed, we say a L percent change in X leads to a (coefficient*L) change in Y.

Question: Do we want to run this analysis for only OAF farmers? If so, adjust the sample accordingly.

logVars <- paste(names(d[grep("log", names(d))]), collapse=" + ")

list2 <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~",  logVars, "+ as.factor(cell)", sep="")), data=d)
  return(mod)
})

list2b <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~",  logVars, sep="")), data=d)
  return(mod)
})

suppressWarnings(
  stargazer(list2, list2b, type="html", 
          title = "2016A Rwanda Soil Health Baseline - Log Agronomic Practice Models",
          covariate.labels = c("Seasons of Fertilizer (log)", "Seasons of Compost (log)", "Seasons of Lime (log)", "Seasons of Fallow (log)"),
          column.labels = c(rep(gsub("m3.","", soilVars),2)),
          dep.var.caption = "",
          dep.var.labels = c("",""),
          add.lines = list(c("Cell FE?", rep("Yes", 5), rep("No", 5))),
          notes = "Includes FE for cell",
          omit=c("cell"), out=paste(od, "rw_baseline_agprac_log.htm", sep="/"))
)

plm.log <- function(x, range){
  
  beta = round(summary(x)$coefficients[range,1],3)
  beta.pval = summary(x)$coefficients[range,4]
  beta.conv = ifelse(beta.pval < 0.01, "***", ifelse(
    beta.pval < 0.05, "**", ifelse(
      beta.pval < 0.1, "*", "")))
  #beta.pval = round(beta.pval, 3)
  outcome = paste(beta, beta.conv, sep = "")
  outcome = c(outcome, unique(round(summary(x)$coefficients[1,1],3)))
  res = data.frame(outcome, stringsAsFactors = F)
  
  return(res)  
}

rw.table15 <- do.call(cbind, lapply(list2, function(x){
  plm.log(x, 2:5)
}))

colnames(rw.table15) <- soilVars
rownames(rw.table15) <- c(names(d)[grep("log", names(d))], "constant")

Table 15 outcomes

rw.table15 <- rw.table15[,c("pH", "Total.C", "Total.N", "m3.Ca", "m3.Mg", "ExAl")]

write.csv(rw.table15, file=paste(od, "rwtable15_reg.csv", sep = "/"),
          row.names = T)

# a 10 percent increase in x leads to B1*.1 change in Y
logTrans <- do.call(cbind, lapply(list2, function(x){
  coeff = x$coefficients[2:5]
  tenPercent = round(coeff*.1, 5)
  names(tenPercent) <- paste("10% increase in ", gsub("log","", names(tenPercent)), " leads to:", sep="")
  return(tenPercent)
}))
colnames(logTrans) <- soilVars

print(kable(logTrans))

	m3.Ca	m3.Mg	pH	Total.N	Total.C	ExAl
10% increase in Fert leads to:	-2.31951	-0.52268	-0.00091	-0.00016	-0.00132	-0.00013
10% increase in Compost leads to:	1.00464	0.17075	0.00180	0.00007	0.00074	-0.00217
10% increase in Lime leads to:	12.28263	2.25366	0.00658	0.00053	0.00676	0.00040
10% increase in Fallow leads to:	-7.04768	-1.33558	-0.01025	0.00011	0.00222	0.00895

Interpretation: When we transform the variables to log, the data starts to tell a more coherent story, at least directionally. If we remove the district FE, the coefficients gain significance.

• Additional seasons of fertilizer use relates to a decrease in soil N and C.
• Seasons of compost and lime have the opposite, positive effect on soil N.
• Unsurprisingly, soil health metrics are poorer for soils on which more fertilizer has been used and lime has not been used.

Tenure with One Acre Fund

Let’s look first at a naive model of One Acre Fund tenure on soil health. Remember: these data are not longitudinal! These data are not longitudinal and reflect farmer selection into One Acre Fund. While these models will try to be both robust and parsimonious, we will inevitabily suffer omitted variable bias due to a lack of an instrument.

list3 <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~",  "total.seasons + as.factor(cell)", sep="")), data=d)
  return(mod)
})

stargazer(list3, type="html", 
          title = "2016A Rwanda Soil Health Baseline - Naive Tenure Models",
          covariate.labels = c("OAF Tenure"),
          dep.var.caption = "",
          dep.var.labels = "",
          column.labels = c(gsub("m3.","", soilVars)),
          notes = "Includes FE for cell",
          omit=c("cell"), out=paste(od, "rw_baseline_tenure.htm", sep="/"))

2016A Rwanda Soil Health Baseline - Naive Tenure Models

	Ca	Mg	pH	Total.N	Total.C	ExAl
	(1)	(2)	(3)	(4)	(5)	(6)
OAF Tenure	-0.817	-0.104	0.002	-0.0002	-0.001	-0.006*
	(2.856)	(0.603)	(0.003)	(0.0002)	(0.003)	(0.003)
Constant	556.919	69.700	4.693***	0.200***	3.370***	1.893***
	(389.276)	(82.243)	(0.373)	(0.021)	(0.392)	(0.422)
Observations	2,443	2,443	2,443	2,443	2,443	2,443
R2	0.550	0.581	0.606	0.500	0.455	0.565
Adjusted R2	0.510	0.544	0.570	0.455	0.406	0.527
Residual Std. Error (df = 2241)	389.234	82.235	0.373	0.021	0.392	0.422
F Statistic (df = 201; 2241)	13.648***	15.480***	17.114***	11.128***	9.300***	14.510***
Note:	p<0.1; p<0.05; p<0.01
	Includes FE for cell

Interpretation: The naive One Acre Fund tenure model suggest that across the board that additional years of 1AF practices have a negative effect on soil health parameters. Let’s combine 1AF tenure with the agronomic practices model above to build a more robust model:

list4 <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~",  logVars, "+ total.seasons + as.factor(cell)", sep="")), data=d)
  return(mod)
})

stargazer(list4, type="html", 
          title = "2016A Rwanda Soil Health Baseline - Ag Practice and Tenure",
          covariate.labels = c("Seasons of Fertilizer (log)", "Seasons of Compost (log)", "Seasons of Lime (log)", "Seasons of Fallow (log)", "OAF Tenure"),
          dep.var.caption = "",
          dep.var.labels = "",
          column.labels = c(gsub("m3.","", soilVars)),
          notes = "Includes FE for cell",
          omit=c("cell"), out=paste(od, "rw_baseline_ag_tenure.htm", sep="/"))

2016A Rwanda Soil Health Baseline - Ag Practice and Tenure

	Ca	Mg	pH	Total.N	Total.C	ExAl
	(1)	(2)	(3)	(4)	(5)	(6)
Seasons of Fertilizer (log)	-27.500**	-6.689**	-0.019	-0.001*	-0.011	0.017
	(13.945)	(2.951)	(0.013)	(0.001)	(0.014)	(0.015)
Seasons of Compost (log)	10.127	1.735	0.018	0.001	0.007	-0.022*
	(12.169)	(2.575)	(0.012)	(0.001)	(0.012)	(0.013)
Seasons of Lime (log)	123.055***	22.614***	0.066**	0.005***	0.068**	0.003
	(32.649)	(6.909)	(0.031)	(0.002)	(0.033)	(0.035)
Seasons of Fallow (log)	-70.840***	-13.479***	-0.103***	0.001	0.022	0.091***
	(15.501)	(3.280)	(0.015)	(0.001)	(0.016)	(0.017)
OAF Tenure	1.921	0.653	0.004	-0.0001	-0.001	-0.008**
	(3.481)	(0.737)	(0.003)	(0.0002)	(0.004)	(0.004)
Constant	535.144	65.394	4.660***	0.198***	3.357***	1.933***
	(386.869)	(81.862)	(0.369)	(0.021)	(0.392)	(0.420)
Observations	2,443	2,443	2,443	2,443	2,443	2,443
R2	0.558	0.587	0.616	0.502	0.456	0.572
Adjusted R2	0.517	0.549	0.581	0.457	0.407	0.533
Residual Std. Error (df = 2237)	386.306	81.743	0.368	0.021	0.392	0.419
F Statistic (df = 205; 2237)	13.771***	15.513***	17.499***	11.012***	9.161***	14.607***
Note:	p<0.1; p<0.05; p<0.01
	Includes FE for cell

Interpretation: Including agronomic practices and 1AF tenure in the same model dampens the magnitude, but not the significance, of 1AF tenure on soil health outcomes.

Agronomic practices - 15B cultivation practices

Thus far we have looked at aggregated historical plot level practices and their effect on soil health. We also asked farmers about their cultivation practices on their plot in the previous season, 15B. We have more precise information for fertilizer, compost and liming practices for the 15B season.

# scale all the field application variables to ares
d$ares <- d$field.size/100
d$fert1.are <- d$field_kg_fert1_15b/d$ares
d$fert2.are <- d$field_kg_fert2_15b/d$ares
d$fert.total.are <- apply(d[,c("fert1.are", "fert2.are")], 1, function(x){
  sum(x, na.rm=T)
})

d$fert.total.are <- ifelse(is.na(d$fert1.are) & is.na(d$fert2.are), NA, d$fert.total.are)

d$compost.are <- d$field_kg_compost_15b/d$ares
d$lime.are <- d$kg_lime_15b/d$ares

intensityVars <- c("fert1.are", "fert2.are",
                   "compost.are", "lime.are")

cor(d[,intensityVars], use="complete.obs")

##             fert1.are fert2.are compost.are  lime.are
## fert1.are   1.0000000 0.9889407   0.5216134 0.8679373
## fert2.are   0.9889407 1.0000000   0.5328874 0.8539114
## compost.are 0.5216134 0.5328874   1.0000000 0.6236616
## lime.are    0.8679373 0.8539114   0.6236616 1.0000000

#table(d$field_15b_fert_t, useNA = 'ifany')
d$field_15b_fert_t <- tolower(d$field_15b_fert_t)
names(d)[names(d)=="v69"] <- "field_15b_fert_t2"
#table(d$field_15b_fert_t2)
d$field_15b_fert_t2 <- tolower(d$field_15b_fert_t2)

d$dap1 <- ifelse(d$field_15b_fert_t=="dap", d$field_kg_fert1_15b, NA)
d$dap1.are <- d$dap1/d$ares
d$npk1 <- ifelse(grepl("npk", d$field_15b_fert_t), d$field_kg_fert1_15b, NA)
d$npk1.are <- d$npk1/d$ares
d$urea1 <- ifelse(d$field_15b_fert_t=="urea", d$field_kg_fert1_15b, NA)
d$urea1.are <- d$urea1/d$ares

d$urea2 <- ifelse(d$field_15b_fert_t2=="urea", d$field_kg_fert2_15b, NA)
d$urea2.are <- d$urea2/d$ares

d$total.urea.are <- apply(d[,c("urea1.are", "urea2.are")], 1, function(x){
  sum(x, na.rm=T)
})
d$total.urea.are <- ifelse(is.na(d$urea1.are) & is.na(d$urea2.are), NA, d$total.urea.are)

Graph the application/are variables

for(i in 1:length(intensityVars)){
  print(
  ggplot(d, aes(x=d[,intensityVars[i]])) + geom_density() +
      labs(x = intensityVars[i], title = intensityVars[i])
  )
}

## Warning: Removed 1800 rows containing non-finite values (stat_density).

## Warning: Removed 2233 rows containing non-finite values (stat_density).

## Warning: Removed 143 rows containing non-finite values (stat_density).

## Warning: Removed 2399 rows containing non-finite values (stat_density).

Conclusion - Take 1: The application rate per are variables are weird. I think it’s because of the field dimensions. I’m going to go back to the field dimensions and check this.

Let’s look at the dimensions of the fields that have large application rates

d[d$fert1.are>10 & !is.na(d$fert1.are), c("field_dim1", "field_dim2", "field.size", "fert1.are")]

##      field_dim1 field_dim2 field.size fert1.are
## 828          48         12        576  15.27778
## 837          10         25        250  35.20000
## 936           5          8         40  25.00000
## 946          25         25        625  14.08000
## 1158          4         20         80  12.50000
## 2380          3          5         15  13.33333

d[d$fert2.are>10 & !is.na(d$fert2.are), c("field_dim1", "field_dim2", "field.size", "fert2.are")]

##     field_dim1 field_dim2 field.size fert2.are
## 828         48         12        576  15.27778
## 830          8         35        280  31.42857
## 837         10         25        250  35.20000
## 877         10          5         50  42.00000

d[d$compost.are>500 & !is.na(d$compost.are), c("field_dim1", "field_dim2", "field.size", "compost.are")]

##      field_dim1 field_dim2 field.size compost.are
## 93           13          8        104    769.2308
## 94            5          4         20   1000.0000
## 181           6          6         36    833.3333
## 205          10          5         50   1000.0000
## 383           2         20         40    750.0000
## 764          26          1         26    807.6923
## 885          15          2         30    666.6667
## 936           5          8         40   1250.0000
## 1128         15          6         90   2223.3333
## 1525         12        123       1476   1507.4526
## 1627         33          3         99    606.0606
## 2366          7         40        280   1714.6429
## 2380          3          5         15    666.6667

# there's a field that is 1 meter wide? Surely not.

d[abs(d$lime.are)>40 & !is.na(d$lime.are), c("field_dim1", "field_dim2", "field.size", "lime.are")]

##      field_dim1 field_dim2 field.size lime.are
## 1101         11         12        132 113.6364
## 1119         25         10        250  60.0000

# how is there a negative quantity of lime?

Table 16 - Kgs of soil ammendments - fertilizer (kg)

# this should be kg of fertilizer used in this field. Compost is off the charts. Convert this to compost per sq meter
plm.t16 <- function(x, range){
  
  beta = round(summary(x)$coefficients[range,1],3)
  beta.pval = round(summary(x)$coefficients[range,4],3)
  beta.conv = ifelse(beta.pval < 0.01, "***", ifelse(
    beta.pval < 0.05, "**", ifelse(
      beta.pval < 0.1, "*", "")))
  #beta.pval = round(beta.pval, 3)
  outcome = paste(beta, " (", beta.pval, ")", sep = "")
  outcome = c(outcome, unique(round(summary(x)$coefficients[1,1],3)))
  res = data.frame(outcome, stringsAsFactors = F)
  
  return(res)  
}


previousSeasonfert <- paste("fert1.are", "as.factor(cell)", sep=" + ")

list5 <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~", previousSeasonfert, sep="")), data=d)
  return(mod)
})

table16 <- do.call(cbind, lapply(list5, function(x){
  plm.t16(x, 2)
}))
colnames(table16) <- soilVars
rownames(table16) <- c("Fertilizer (kg/are)", "constant")

table16 <- table16[,c("pH", "Total.C", "Total.N", "m3.Ca", "m3.Mg", "ExAl")]
write.csv(table16, file=paste(od, "rwtable16_fert.csv", sep = "/"),
          row.names = T)

Table 16 - all fertilizer used

previousSeasonfertAll <- paste("fert.total.are", "as.factor(cell)", sep=" + ")

list5a <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~", previousSeasonfertAll, sep="")), data=d)
  return(mod)
})

table16a <- do.call(cbind, lapply(list5a, function(x){
  plm.t16(x, 2)
}))
colnames(table16a) <- soilVars
rownames(table16a) <- c("Fertilizer (all) (kg/are)", "constant")

table16a <- table16a[,c("pH", "Total.C", "Total.N", "m3.Ca", "m3.Mg", "ExAl")]
write.csv(table16a, file=paste(od, "rwtable16_fert_all.csv", sep = "/"),
          row.names = T)

Table 16 - Kgs of soil ammendments - compost (kg)

# these objects have the same name as the fertilizer objects for simplicity. Run
# from the top to avoid overwriting issues.
previousSeasoncompost <- paste("compost.are", "as.factor(cell)", sep=" + ")
list5b <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~", previousSeasoncompost, sep="")), data=d)
  return(mod)
})

table16b <- do.call(cbind, lapply(list5b, function(x){
  plm.t16(x, 2)
}))
colnames(table16b) <- soilVars
rownames(table16b) <- c("Compost (kg/are)", "constant")

table16b <- table16b[,c("pH", "Total.C", "Total.N", "m3.Ca", "m3.Mg", "ExAl")]
write.csv(table16b, file=paste(od, "rwtable16_compost.csv", sep = "/"),
          row.names = T)

Table 16 - Kgs of soil ammendments - lime (kg)

# these objects have the same name as the fertilizer objects for simplicity. Run
# from the top to avoid overwriting issues.
previousSeasonlime <- paste("lime.are", "as.factor(cell)", sep=" + ")
list5c <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~", previousSeasonlime, sep="")), data=d)
  return(mod)
})

table16c <- do.call(cbind, lapply(list5c, function(x){
  plm.t16(x, 2)
}))
colnames(table16c) <- soilVars
rownames(table16c) <- c("Lime (kg/are)", "constant")

table16c <- table16c[,c("pH", "Total.C", "Total.N", "m3.Ca", "m3.Mg", "ExAl")]
write.csv(table16c, file=paste(od, "rwtable16_lime.csv", sep = "/"),
          row.names = T)

Table 16 - types of fertilizer - DAP (kg)

previousSeasonDAP <- paste("dap1.are", "as.factor(cell)", sep=" + ")
list5d <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~", previousSeasonDAP, sep="")), data=d)
  return(mod)
})

table16d <- do.call(cbind, lapply(list5d, function(x){
  plm.t16(x, 2)
}))
colnames(table16d) <- soilVars
rownames(table16d) <- c("DAP (kg/are)", "constant")

table16d <- table16d[,c("pH", "Total.C", "Total.N", "m3.Ca", "m3.Mg", "ExAl")]
write.csv(table16d, file=paste(od, "rwtable16_dap.csv", sep = "/"),
          row.names = T)

Table 16 - types of fertilizer - NPK (kg)

previousSeasonNPK <- paste("npk1.are", "as.factor(cell)", sep=" + ")
list5e <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~", previousSeasonNPK, sep="")), data=d)
  return(mod)
})

table16e <- do.call(cbind, lapply(list5e, function(x){
  plm.t16(x, 2)
}))
colnames(table16e) <- soilVars
rownames(table16e) <- c("NPK (kg/are)", "constant")

table16e <- table16e[,c("pH", "Total.C", "Total.N", "m3.Ca", "m3.Mg", "ExAl")]
write.csv(table16e, file=paste(od, "rwtable16_npk.csv", sep = "/"),
          row.names = T)

Table 16 - types of fertilizer - Urea (kg)

previousSeasonUrea <- paste("total.urea.are", "as.factor(cell)", sep=" + ")
list5f <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~", previousSeasonUrea, sep="")), data=d)
  return(mod)
})

table16f <- do.call(cbind, lapply(list5f, function(x){
  plm.t16(x, 2)
}))
colnames(table16f) <- soilVars
rownames(table16f) <- c("Urea (kg/are)", "constant")

table16f <- table16f[,c("pH", "Total.C", "Total.N", "m3.Ca", "m3.Mg", "ExAl")]
write.csv(table16f, file=paste(od, "rwtable16_urea.csv", sep = "/"),
          row.names = T)

stargazer(list5, type="html", 
          title = "2016A Rwanda Soil Health Baseline - 15B practices",
          covariate.labels = c("Fertilizer Rate (log)"),
          dep.var.caption = "",
          dep.var.labels = "",
          column.labels = c(gsub("m3.","", soilVars)),
          notes = "Includes FE for cell",
          omit=c("cell"), out=paste(od, "rw_baseline_15b_ag.htm", sep="/"))

Farmer fertility perception

Let’s look at farmer perceived fertility as a predictor of soil health. We’ll set ‘same fertility’ as the reference category.

d$fertility_qual <- relevel(as.factor(d$general_field_infocompare_fertil), ref="same")

list6 <- lapply(soilVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~", "+ fertility_qual + as.factor(cell)", sep="")), data=d)
  return(mod)
})


stargazer(list6, type="html", 
          title = "2016A Rwanda Soil Health Baseline - Farmer Perceived Fertility",
          covariate.labels = c("Farmer Opinion - Less Fertile",
                               "Farmer Opinion - More Fertile"),
          dep.var.caption = "",
          dep.var.labels = "",
          column.labels = c(gsub("m3.","", soilVars)),
          notes = "Reference category is same fertility as other fields. Includes FE for cell",
          notes.align = "l",
          omit=c("cell"), out=paste(od, "rw_baseline_fertility_qual.htm", sep="/"))

2016A Rwanda Soil Health Baseline - Farmer Perceived Fertility

	Ca	Mg	pH	Total.N	Total.C	ExAl
	(1)	(2)	(3)	(4)	(5)	(6)
Farmer Opinion - Less Fertile	-110.261***	-19.192***	-0.124***	-0.0002	0.003	0.139***
	(21.695)	(4.593)	(0.021)	(0.001)	(0.022)	(0.024)
Farmer Opinion - More Fertile	7.185	1.252	-0.005	-0.001	-0.001	0.001
	(23.953)	(5.071)	(0.023)	(0.001)	(0.024)	(0.026)
Constant	665.546*	88.684	4.822***	0.200***	3.364***	1.742***
	(387.433)	(82.024)	(0.370)	(0.021)	(0.393)	(0.420)
Observations	2,443	2,443	2,443	2,443	2,443	2,443
R2	0.556	0.585	0.612	0.499	0.455	0.572
Adjusted R2	0.516	0.548	0.577	0.454	0.406	0.533
Residual Std. Error (df = 2240)	386.825	81.895	0.370	0.021	0.392	0.419
F Statistic (df = 202; 2240)	13.893***	15.629***	17.488***	11.063***	9.248***	14.820***
Note:	p<0.1; p<0.05; p<0.01
	Reference category is same fertility as other fields. Includes FE for cell

Interpretation: Farmers understand their fields well. Their categorization of which field are more and less fertile corresponds to our quantified measures of soil health. The only features farmers don’t seem to get correct are nitrogen and carbon. The nitrogen and carbon levels are indistinguishable in ‘low fertility’ fields relative to the fields deemed to be the ‘same fertility.’ Reminder: We need to remember that farmers are only evaluting one of their fields thus we are not able to account for the quality of the farmer in assessing his/her fields.

Deeper dive into farmer perception of fertility

• Look at wet chem values for all soil features by farmer perception
• Run PCA on all soil attributes to see which principle components predict soil fertility
• Consider adding additional control variables into model (like Ca as control for pH perception model)

# merge wetChem in with d
names(wetChem)[2:21] <- paste("wet.", names(wetChem)[2:21], sep = "")
d <- left_join(d, wetChem, by="SSN")

## Warning in left_join_impl(x, y, by$x, by$y, suffix$x, suffix$y): joining
## factor and character vector, coercing into character vector

wetVars <- names(d)[grep("wet.", names(d))]
for(i in 1:length(wetVars)){
  print(
  ggplot(data=d, aes(x=as.factor(client), y=d[,wetVars[i]])) + 
    geom_boxplot() +
    labs(x="Tubura Farmer", y=wetVars[i], title = paste("RW baseline wet chem - ", wetVars[i], sep = ""))
  )
}

## Warning: Removed 2206 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2206 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

## Warning: Removed 2199 rows containing non-finite values (stat_boxplot).

list7 <- lapply(wetVars, function(x){
  mod <- lm(as.formula(paste("d[,x] ~", "+ fertility_qual + as.factor(cell)", sep="")), data=d)
  return(mod)
})

suppressWarnings(
stargazer(list7, type="html", 
          title = "2016A Rwanda Soil Health Baseline - Farmer Perceived Fertility (wet chem)",
          covariate.labels = c("Farmer Opinion - Less Fertile",
                               "Farmer Opinion - More Fertile"),
          dep.var.caption = "",
          dep.var.labels = "",
          column.labels = c(gsub("wet.","", wetVars)),
          notes = "Reference category is same fertility as other fields. Includes FE for cell",
          notes.align = "l",
          omit=c("cell"), out=paste(od, "rw_baseline_fertility_qual_wet.htm", sep="/"))
)

2016A Rwanda Soil Health Baseline - Farmer Perceived Fertility (wet chem)

	pH	EC..S.	P	K	Ca	Mg	S	Na	Fe	Mn	B	Cu	Zn	C.E.C	TN	C.N	Exch..Acidity	Acid.Saturation	Exch.Al	OC
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)	(14)	(15)	(16)	(17)	(18)	(19)	(20)
Farmer Opinion - Less Fertile	-0.227	1.930	-9.111	-44.922	-23.678	-22.846	2.274	-0.079	22.320	-29.748	-0.064	-0.317	-0.224	0.138	0.004	-0.277	0.343**	9.657**	0.268**	0.153
	(0.147)	(11.629)	(19.670)	(28.712)	(150.522)	(36.856)	(2.236)	(2.929)	(17.023)	(21.301)	(0.222)	(0.337)	(0.764)	(1.126)	(0.008)	(0.627)	(0.160)	(3.686)	(0.124)	(0.135)
Farmer Opinion - More Fertile	-0.109	-7.216	-16.525	-69.198**	-187.493	-57.688	2.223	-0.854	-12.790	-10.391	-0.069	-0.278	-1.040	-1.369	-0.016*	0.287	0.125	3.831	0.091	-0.192
	(0.174)	(13.775)	(23.740)	(34.653)	(181.669)	(44.482)	(2.699)	(3.535)	(20.546)	(25.708)	(0.268)	(0.407)	(0.922)	(1.359)	(0.010)	(0.757)	(0.193)	(4.449)	(0.150)	(0.163)
Constant	6.225***	127.000***	8.910	213.000***	1,745.000***	435.000***	15.700**	24.300***	61.450	194.000***	0.415	2.670***	4.415**	15.650***	0.195***	11.750***	0.140	1.100	0.023	2.335***
	(0.406)	(32.067)	(55.599)	(81.159)	(425.475)	(104.178)	(6.321)	(8.278)	(48.119)	(60.210)	(0.627)	(0.953)	(2.160)	(3.182)	(0.023)	(1.772)	(0.451)	(10.419)	(0.351)	(0.383)
Observations	237	237	244	244	244	244	244	244	244	244	244	244	244	244	244	244	244	244	244	244
R2	0.742	0.580	0.702	0.653	0.760	0.755	0.729	0.705	0.769	0.688	0.525	0.693	0.540	0.752	0.773	0.671	0.775	0.777	0.797	0.731
Adjusted R2	0.425	0.064	0.348	0.240	0.475	0.463	0.407	0.354	0.494	0.318	-0.039	0.329	-0.006	0.457	0.504	0.280	0.506	0.511	0.555	0.412
Residual Std. Error	0.574 (df = 106)	45.349 (df = 106)	78.629 (df = 111)	114.776 (df = 111)	601.713 (df = 111)	147.330 (df = 111)	8.940 (df = 111)	11.707 (df = 111)	68.051 (df = 111)	85.150 (df = 111)	0.887 (df = 111)	1.347 (df = 111)	3.055 (df = 111)	4.501 (df = 111)	0.033 (df = 111)	2.506 (df = 111)	0.638 (df = 111)	14.735 (df = 111)	0.496 (df = 111)	0.541 (df = 111)
F Statistic	2.344*** (df = 130; 106)	1.125 (df = 130; 106)	1.983*** (df = 132; 111)	1.580*** (df = 132; 111)	2.664*** (df = 132; 111)	2.586*** (df = 132; 111)	2.264*** (df = 132; 111)	2.010*** (df = 132; 111)	2.796*** (df = 132; 111)	1.857*** (df = 132; 111)	0.930 (df = 132; 111)	1.902*** (df = 132; 111)	0.988 (df = 132; 111)	2.551*** (df = 132; 111)	2.867*** (df = 132; 111)	1.715*** (df = 132; 111)	2.888*** (df = 132; 111)	2.923*** (df = 132; 111)	3.293*** (df = 132; 111)	2.290*** (df = 132; 111)
Note:	p<0.1; p<0.05; p<0.01
	Reference category is same fertility as other fields. Includes FE for cell

Interpretation: Our sample size decreases considerably when we look only at wet chemistry results. Thus, we do not see the same signficant relationships we saw between farmer perceived fertility and soil characteristics we saw when looking at the predicted values.

Principal Component Analysis (predicted values)

pca1 <- prcomp(d[,soilVars], scale=TRUE, center=TRUE)
print(pca1)

## Standard deviations:
## [1] 1.8130261 1.4406298 0.5220816 0.4155603 0.3485684 0.2660126
## 
## Rotation:
##                PC1         PC2         PC3         PC4         PC5
## m3.Ca    0.4639924 -0.32390991  0.05174966 -0.49857472  0.10898630
## m3.Mg    0.4680317 -0.22493445  0.73859014  0.35920491  0.02360238
## pH       0.5252283  0.01946359 -0.31205314 -0.40917507 -0.29772280
## Total.N -0.1011532 -0.65539834 -0.27462802  0.04606937  0.62468884
## Total.C -0.1777593 -0.62959112 -0.12420105  0.20482827 -0.71038388
## ExAl    -0.4979581 -0.13481869  0.51340332 -0.64101002 -0.06359753
##                 PC6
## m3.Ca    0.64549254
## m3.Mg   -0.23505354
## pH      -0.60853607
## Total.N -0.30404269
## Total.C  0.09995442
## ExAl    -0.23524504

plot(pca1)

pca1$rotation

##                PC1         PC2         PC3         PC4         PC5
## m3.Ca    0.4639924 -0.32390991  0.05174966 -0.49857472  0.10898630
## m3.Mg    0.4680317 -0.22493445  0.73859014  0.35920491  0.02360238
## pH       0.5252283  0.01946359 -0.31205314 -0.40917507 -0.29772280
## Total.N -0.1011532 -0.65539834 -0.27462802  0.04606937  0.62468884
## Total.C -0.1777593 -0.62959112 -0.12420105  0.20482827 -0.71038388
## ExAl    -0.4979581 -0.13481869  0.51340332 -0.64101002 -0.06359753
##                 PC6
## m3.Ca    0.64549254
## m3.Mg   -0.23505354
## pH      -0.60853607
## Total.N -0.30404269
## Total.C  0.09995442
## ExAl    -0.23524504

The first principal component is composed primarily of soil pH related variables, Ca, Mg, and pH. The second capture N and C. These groupings (pH grouping and C and N) are not all that surprising given that our predicted soil variable set is fairly limited.

If the variables have the same sign that indicates that they are positively correlated in the principal component. In the first principal component, we see Ca, Mg and pH loading in the same direction. In the second principal component, we see N and C moving in the same direction. Ca and Mg are also associated in the same direction but to a lesser extent.

ggplot(as.data.frame(pca1$x),aes(x=PC1,y=PC2, color=d$fertility_qual)) + geom_point() + 
  labs(title = "PCA of soil attributes and farmer perceived soil fertility",
       x = "First PCA", y= "Second PCA", color="Field Quality")

When we layer farmer perceived soil fertility on top of the principal component scatter plot, no clear pattern emergres. Fields with the same fertilitiy, less and more fertility are indistinguishable by the principal components. Thus, there is not a clear profile for farmer identified healthier soil or weaker soil based on principal components.

Geographic Descriptors - PCA for soil profiles

This section will build on the principal component work above and look at improving understanding of local context to inform local adaptation. Here we will also construct soil profiles to simplify the scaling of promising products and practices to targeted locations. We don’t have a full suite of predictors so we can’t look at a comprehensive soil profile.

ggplot(as.data.frame(pca1$x),aes(x=PC1,y=PC2, color=d$district)) + geom_point() + 
  labs(title = "PCA of soil attributes and district",
       x = "First PCA", y= "Second PCA", color="District")

When we color the figure by district, we start to see a pattern emerge. However, there are too many points to clearly detect where all districts fall. Let’s instead look at the data by AEZ.

ggplot(as.data.frame(pca1$x),aes(x=PC1,y=PC2, color=d$aez)) + geom_point() + 
  labs(title = "PCA of soil attributes and AEZ",
       x = "First PCA - Ca/Mg/pH", y= "Second PCA - N and C", color="AEZ")

Coloring the points by AEZ, we see a much clearer trend. The east is to the right of our graphic, then the central plateau followed by lake Kive and then Congo Nile in green on the left. What does this mean in terms of actual soil features? Let’s look at the figure but with the variable associations layered on top. See here for documentation

library(pca3d)
pca2d( pca1, biplot= TRUE, shape= 19, col= "black"  )

It appears that the right side of the graph, the eastern AEZ, is associated with pH, Mg and Ca. This indicates that as the first principal component increases, so does the level of pH, Ca and Mg. Conversely, total N and C increase as the second principal component decreases. In terms of the AEZ figure above, this suggest that the Congo Nile AEZ has higher levels of N and C. Let’s test these hypotheses with a simple summary table:

print(kable(aggregate(d[,soilVars], by=list(d$aez), function(x){
  round(mean(x, na.rm=T),5)
})))

Group.1	m3.Ca	m3.Mg	pH	Total.N	Total.C	ExAl
Central Plateau	850.4809	192.4942	5.64775	0.14560	1.90045	0.40840
Congo Nile	479.9955	124.0777	5.02279	0.16663	2.37550	1.17301
East	1369.4056	276.0773	6.01196	0.16728	2.13170	0.18844
Lake Kivu	798.7319	236.6825	5.47968	0.14884	2.05310	0.55349

Confirmed! This likely also suggests that Congo Nile is at a higher altitude than the surrounding areas and that eastern Rwanda has relatively less weathered soils compared to western.

Consequence for trial placement: The Rwanda program is already blocking trials by AEZ but these data confirm that AEZ reflects meaningfu soil variation and thus captures key growing conditions for Rwandan farmers. Blocking trials by AEZ in Rwanda will enable us to evaluate trial hypotheses in more neutral pH ranges and higher N and C conditions.

Principal Component Analysis (wet chemistry values)

wetVal <- d[complete.cases(d[,wetVars]),]

pca2 <- prcomp(wetVal[,c("wet.C.E.C", "wet.pH", "wet.Mg", "wet.Ca")], center=TRUE, scale=TRUE)

pca2 <- prcomp(wetVal[, wetVars], center=TRUE, scale=TRUE)

#plot(pca2)
ggplot(as.data.frame(pca2$x),aes(x=PC1,y=PC2, color=as.factor(wetVal$aez))) + geom_point() +
  labs(title = "Wet Chem PCA with AEZ", x = "First PC", y="Second PC",
       color="AEZ")

#pca2$rotation

pca2d(pca2, biplot= TRUE, shape= 19, col= "black")

# put pca2$x PC1 in the main data to run the fertility perception model.
#mod8 <- lm(
suppressWarnings(
stargazer(list7, type="html", 
          title = "2016A Rwanda Soil Health Baseline - Farmer Perceived Fertility (wet chem)",
          covariate.labels = c("Farmer Opinion - Less Fertile",
                               "Farmer Opinion - More Fertile"),
          dep.var.caption = "",
          dep.var.labels = "",
          column.labels = c(gsub("wet.","", wetVars)),
          notes = "Reference category is same fertility as other fields. Includes FE for cell",
          notes.align = "l",
          omit=c("cell"), out=paste(od, "rw_baseline_fertility_qual_wet.htm", sep="/"))
)

2016A Rwanda Soil Health Baseline - Farmer Perceived Fertility (wet chem)

	pH	EC..S.	P	K	Ca	Mg	S	Na	Fe	Mn	B	Cu	Zn	C.E.C	TN	C.N	Exch..Acidity	Acid.Saturation	Exch.Al	OC
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)	(14)	(15)	(16)	(17)	(18)	(19)	(20)
Farmer Opinion - Less Fertile	-0.227	1.930	-9.111	-44.922	-23.678	-22.846	2.274	-0.079	22.320	-29.748	-0.064	-0.317	-0.224	0.138	0.004	-0.277	0.343**	9.657**	0.268**	0.153
	(0.147)	(11.629)	(19.670)	(28.712)	(150.522)	(36.856)	(2.236)	(2.929)	(17.023)	(21.301)	(0.222)	(0.337)	(0.764)	(1.126)	(0.008)	(0.627)	(0.160)	(3.686)	(0.124)	(0.135)
Farmer Opinion - More Fertile	-0.109	-7.216	-16.525	-69.198**	-187.493	-57.688	2.223	-0.854	-12.790	-10.391	-0.069	-0.278	-1.040	-1.369	-0.016*	0.287	0.125	3.831	0.091	-0.192
	(0.174)	(13.775)	(23.740)	(34.653)	(181.669)	(44.482)	(2.699)	(3.535)	(20.546)	(25.708)	(0.268)	(0.407)	(0.922)	(1.359)	(0.010)	(0.757)	(0.193)	(4.449)	(0.150)	(0.163)
Constant	6.225***	127.000***	8.910	213.000***	1,745.000***	435.000***	15.700**	24.300***	61.450	194.000***	0.415	2.670***	4.415**	15.650***	0.195***	11.750***	0.140	1.100	0.023	2.335***
	(0.406)	(32.067)	(55.599)	(81.159)	(425.475)	(104.178)	(6.321)	(8.278)	(48.119)	(60.210)	(0.627)	(0.953)	(2.160)	(3.182)	(0.023)	(1.772)	(0.451)	(10.419)	(0.351)	(0.383)
Observations	237	237	244	244	244	244	244	244	244	244	244	244	244	244	244	244	244	244	244	244
R2	0.742	0.580	0.702	0.653	0.760	0.755	0.729	0.705	0.769	0.688	0.525	0.693	0.540	0.752	0.773	0.671	0.775	0.777	0.797	0.731
Adjusted R2	0.425	0.064	0.348	0.240	0.475	0.463	0.407	0.354	0.494	0.318	-0.039	0.329	-0.006	0.457	0.504	0.280	0.506	0.511	0.555	0.412
Residual Std. Error	0.574 (df = 106)	45.349 (df = 106)	78.629 (df = 111)	114.776 (df = 111)	601.713 (df = 111)	147.330 (df = 111)	8.940 (df = 111)	11.707 (df = 111)	68.051 (df = 111)	85.150 (df = 111)	0.887 (df = 111)	1.347 (df = 111)	3.055 (df = 111)	4.501 (df = 111)	0.033 (df = 111)	2.506 (df = 111)	0.638 (df = 111)	14.735 (df = 111)	0.496 (df = 111)	0.541 (df = 111)
F Statistic	2.344*** (df = 130; 106)	1.125 (df = 130; 106)	1.983*** (df = 132; 111)	1.580*** (df = 132; 111)	2.664*** (df = 132; 111)	2.586*** (df = 132; 111)	2.264*** (df = 132; 111)	2.010*** (df = 132; 111)	2.796*** (df = 132; 111)	1.857*** (df = 132; 111)	0.930 (df = 132; 111)	1.902*** (df = 132; 111)	0.988 (df = 132; 111)	2.551*** (df = 132; 111)	2.867*** (df = 132; 111)	1.715*** (df = 132; 111)	2.888*** (df = 132; 111)	2.923*** (df = 132; 111)	3.293*** (df = 132; 111)	2.290*** (df = 132; 111)
Note:	p<0.1; p<0.05; p<0.01
	Reference category is same fertility as other fields. Includes FE for cell

District and cell level summaries of soil and managment practices

dist.sum <- aggregate(d[,out.list], by=list(d$district), function(x){
  return(c(
    paste(
      round(mean(x, na.rm=T),3), " (", round(median(x, na.rm=T),3), ")", 
      " (", round(sd(x, na.rm=T),2), ")", sep = ""
    )
    ))
})

write.csv(dist.sum, file=paste(od, "district covariate summary.csv", sep = "/"))

cell.sum <- aggregate(d[,out.list], by=list(d$cell), function(x){
  return(c(
    paste(
      round(mean(x, na.rm=T),3), " (", round(median(x, na.rm=T),3), ")", 
      " (", round(sd(x, na.rm=T),2), ")", sep = ""
    )
    ))
})

write.csv(cell.sum, file=paste(od, "cell covariate summary.csv", sep = "/"))

Female farmers farming poorer soils?

remove <- "female"
genderBalance <- out.list[!out.list %in% remove]

equal <- do.call(rbind, lapply(genderBalance, function(x){
    
    out <- t.test(d[,x] ~ d[,"female"], data=d)
    tab <- data.frame(out[[5]][[2]],out[[5]][[1]], out[3])
    tab[,1:2] <- round(tab[,1:2],3)
    names(tab) <- c(names(out[[5]]), "pvalue")
    #tab[,3] <- p.adjust(tab[,3], method="holm")
    #tab[,3] <- ifelse(tab[,3] < 0.001, "< 0.001", round(tab[,3],3))
    #print(tab)
    return(tab)
  
}))

rownames(equal) <- NULL

# order variables 
equal$pvalue <- p.adjust(equal$pvalue, method="fdr")
rownames(equal) <- genderBalance
equal <- equal[order(equal$pvalue),]

equal$pvalue <- ifelse(equal$pvalue < 0.001, "< 0.001", round(equal$pvalue,3))
colnames(equal) <- c("Male Farmers","Female Farmers", "p-value")    

write.csv(equal, file=paste(od, "female farmer status.csv", sep = "/"), row.names=T)

Propensity Score Matching

We need to do a more rigorous job of accounting for differences between Tubura farmers and identified control farmers. Execute propensity score matching (PSM) to identify control farmers that overlap with Tubura farmers with regard to their likelihood of being a Tubura farmer.

psmVars <- paste(c("female", "age", "hhsize", "total.seasons",
                   "cows", "goats", "chickens", "pigs", "sheep"),
                   collapse=" + ")

reg <- glm(as.formula(paste("client ~", psmVars, sep="")), 
           family= binomial(link="logit"), data=d)
summary(reg)    

## 
## Call:
## glm(formula = as.formula(paste("client ~", psmVars, sep = "")), 
##     family = binomial(link = "logit"), data = d)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -3.5522  -0.8456   0.0376   0.9250   2.0863  
## 
## Coefficients:
##                Estimate Std. Error z value Pr(>|z|)    
## (Intercept)    0.408940   0.203191   2.013   0.0442 *  
## female        -0.749686   0.097393  -7.698 1.39e-14 ***
## age           -0.024394   0.003378  -7.221 5.16e-13 ***
## hhsize         0.037275   0.022980   1.622   0.1048    
## total.seasons  0.530099   0.028803  18.404  < 2e-16 ***
## cows           0.007891   0.017934   0.440   0.6599    
## goats         -0.004426   0.030020  -0.147   0.8828    
## chickens       0.023011   0.017794   1.293   0.1959    
## pigs           0.052035   0.057838   0.900   0.3683    
## sheep          0.053211   0.070949   0.750   0.4533    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 3386.7  on 2442  degrees of freedom
## Residual deviance: 2566.6  on 2433  degrees of freedom
## AIC: 2586.6
## 
## Number of Fisher Scoring iterations: 5

# summarize predicted probabilities
pr <- data.frame(pr_score = predict(reg, type='response'), treat = d$client)

# graph
psmGraph <- ggplot() + geom_histogram(data=subset(pr, pr$treat==1), aes(x = pr_score, y=..count.., fill=as.factor(treat)), bins=80, position = "identity") +
    geom_histogram(data=subset(pr, pr$treat==0), aes(x=pr_score, y=-..count.., fill=as.factor(treat)), bins=80, position = "identity") +
    scale_y_continuous(limits=c(-150,150)) + 
  labs(title ="PSM score overlap", x = "PSM score", y="Farmer count",
       fill="Tubura/Control")

print(psmGraph)

pdf(file=paste(od, "rw_baseline_psm_overlap.pdf", sep="/"), height=8.5, width=11)
print(psmGraph)
dev.off()

## quartz_off_screen 
##                 2

Interpretation We have some overlap but it’s clear that Tubura farmers occupy a different range than the identified control farmers. Let’s continue with the PSM matching process but restrict ourselves to Tubura and control farmers that meet a certain PSM matching radius.

Notes on PSM process:

We have to indicate a variable for matching. I’m choosing pH as we know it to be an issue in most of our operating areas and addressing soil acidity has numerous residual benefits to soil health. I’ll want to do this for all soil outcomes, however.

• As a next step, I’d like to identify the matched subset and then regress Tubura tenure on soil outcomes. Does that make sense?
• I chose a caliper of 0.25. I should review this process with Maya to make certain I’m following best practice. I sort of just pulled that sd figure out of the sky.
• An initial check of the match quality using a common model indicates that the matches are poor.

# PSM prep
tr <- cbind(d$client)
x <- d[, unlist(strsplit(psmVars, " + ", fixed=T))]
y <- soilVars


# PSM
set.seed(20161102)
m <- lapply(y, function(response){

  suppressWarnings(
  mod <- Match(Y = d[,response], Tr = tr, X = reg$fitted, ties=FALSE, replace=FALSE, caliper=0.25, estimand = "ATE")    
  )
  matchRes <- MatchBalance(tr ~ d[,response], match.out=mod, nboots=500, data=d, print.level = 0)
  return(list(mod, matchRes))
})
#lapply(m, summary)

Now check the naive model approach for PSM balance.

matchRes <- do.call(rbind, lapply(1:length(m), function(model){
  val <- as.data.frame(cbind(
    standard.diff=m[[model]][[2]]$AfterMatching[[1]]$sdiff, 
    var.ratio = m[[model]][[2]]$AfterMatching[[1]]$var.ratio,
    sdiff.adj = m[[model]][[2]]$AfterMatching[[1]]$sdiff/100))
  return(val)
}))
rownames(matchRes) <- y

print(kable(matchRes))

	standard.diff	var.ratio	sdiff.adj
m3.Ca	-1.5998283	0.8574354	-0.0159983
m3.Mg	-12.0950035	0.7265264	-0.1209500
pH	6.4227217	1.0777240	0.0642272
Total.N	-2.1315722	0.9573257	-0.0213157
Total.C	-2.1472466	1.0490227	-0.0214725
ExAl	-0.8045446	1.0180995	-0.0080454

Interpretation: We want to see standard mean differences less than the absolute value of 0.25 (or 0.1 if we’re being conservative) and variance ratios close to 1 but certainly between 0.5 and 2.

According to the CRAN summary, sdiff is the standardized difference between the treatment and control units multiplied by 100. If I divide by 100, the values come much closer to reasonable value.

Let the PSM models vary

The common model approach doesn’t seem to be working for any of the variables. I’m going to rework the modeling approach so we can fit different models for each outcome upon which we’re trying to match.

d$age2 <- d$age^2
d$hhsize_age <- d$hhsize*d$age
d$hhsize2 <- d$hhsize^2

coreVars = c("female", "age", "hhsize", "own", "as.factor(cell)", "cows", "goats", "chickens", "pigs", "sheep")

psmList <- list(
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="m3.Ca"),
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="m3.Mg"),
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="pH"),
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="Total.N"),
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="Total.C"),
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="ExAl"),
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="n_season_fert"),
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="n_season_compost"),
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="n_seasons_leg_1"),
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="n_season_fallow"),
  list(tr = "client",
       psmVars = paste(coreVars,
                   collapse=" + "),
       y="logFert"),
  list(tr = "client",
       psmVars = paste(c(coreVars, "age2",
                   "hhsize2"),
                   collapse=" + "),
       y="n_seasons_leg_2"),
  list(tr = "client",
       psmVars = paste(c(coreVars, "age2",
                   "hhsize2"),
                   collapse=" + "),
       y="n_season_lime"),
  list(tr = "client",
       psmVars = paste(c("female", "age", "hhsize", "own",
                         "cows", "goats", "chickens", "pigs", "sheep",
                         "as.factor(district)", "age2", "hhsize2",
                         "hhsize_age"),
                   collapse=" + "),
       y="fert1.are"),
  # list(tr = "client",
  #      psmVars = paste(c("female", "age", "hhsize","as.factor(district)",
  #                  "cows", "goats", "chickens", "pigs", "sheep","age2", "hhsize2",
  #                        "hhsize_age"),
  #                  collapse=" + "),
  #      y="fert2.are"),
  list(tr = "client",
       psmVars = paste(c("female", "age", "hhsize","as.factor(district)",
                   "cows", "goats", "chickens", "pigs", "sheep","age2", "hhsize2",
                         "hhsize_age"),
                   collapse=" + "),
       y="compost.are")
  
)

# fertilizer application in 15B has missing values at the cell level so i include
# a district level control instead. We don't get a good fit with the current model.
# adjust!
# aggregate(d$fert1.are, by=list(d$client, d$cell), FUN=mean, na.rm=T)
# lime.are is also being finicky



# PSM
set.seed(20161102)
m <- lapply(psmList, function(listInput){

  naCheck <- unlist(strsplit(gsub("\\+", "", as.vector(listInput$psmVars))," ", fixed=TRUE))
  naCheck <- naCheck[-which(naCheck=="")]
  
  # keep complete cases of outcome variable
  k <- d[complete.cases(d[,listInput$y]),]
  k <- k[complete.cases(k[,listInput$y]),]
  
  # run glm regression:
  reg <- glm(as.formula(paste(listInput$tr, "~", listInput$psmVars, sep="")),  family= binomial(link="logit"), data=k)
  
  suppressWarnings(
  mod <- Match(Y = k[,listInput$y], Tr = k[,listInput$tr], X = reg$fitted, ties=FALSE, replace=FALSE, caliper=0.25, estimand = "ATE")   
  )
  matchRes <- MatchBalance(k[,listInput$tr] ~ k[,listInput$y], match.out=mod, nboots=500, data=k, print.level = 0)
  #print(listInput$y)
  return(list(mod, matchRes))
  
})

The models can now vary by outcome. Let’s see if we can improve our results.

matchRes <- do.call(rbind, lapply(1:length(m), function(model){
  val <- as.data.frame(cbind(
    standard.diff=m[[model]][[2]]$AfterMatching[[1]]$sdiff, 
    var.ratio = m[[model]][[2]]$AfterMatching[[1]]$var.ratio,
    sdiff.adj = m[[model]][[2]]$AfterMatching[[1]]$sdiff/100))
  return(val)
}))

namesInput <- NULL
for(i in 1:length(psmList)){
  namesInput[i] <- psmList[[i]]$y
}
rownames(matchRes) <- namesInput

print(kable(matchRes))

	standard.diff	var.ratio	sdiff.adj
m3.Ca	-4.7500746	0.9005991	-0.0475007
m3.Mg	-6.3887895	0.9215206	-0.0638879
pH	-2.9860193	1.0181086	-0.0298602
Total.N	-6.0596975	0.9986588	-0.0605970
Total.C	-2.3798262	1.0594167	-0.0237983
ExAl	4.1895526	1.0408160	0.0418955
n_season_fert	78.6639739	3.3582015	0.7866397
n_season_compost	9.4102562	0.9632805	0.0941026
n_seasons_leg_1	5.4860055	0.8773639	0.0548601
n_season_fallow	6.2184929	1.2947263	0.0621849
logFert	102.5919067	1.7352434	1.0259191
n_seasons_leg_2	-12.0195420	0.9363917	-0.1201954
n_season_lime	14.0155855	1.5306613	0.1401559
fert1.are	0.5627194	1.3655780	0.0056272
compost.are	8.5920038	1.3089388	0.0859200

Interpretation If I divide the standardized mean differences by 100, we meet the balance criteria of the standardized mean difference being close to 0 and the variance being close to 1. Let’s print out the model results to see how Tubura and control farmers compare on key soil meterics. These results should supercede the naive balance tables presented above.

We achieve acceptable balance for the soil attributes but we don’t for seasons of fertilizer use. This is isn’t entirely unexpected given that Tubura’s primary service is providing fertilizer inputs and training.

coefTable <- do.call(rbind, lapply(1:length(m), function(model){
  beta = round(m[[model]][[1]]$est.noadj,3)
  mean.Tr = round(m[[model]][[2]]$AfterMatching[[1]][[3]], 2)
  mean.Co = round(m[[model]][[2]]$AfterMatching[[1]][[4]], 2)
  pval = m[[model]][[2]]$AfterMatching[[1]][[10]][[3]] # p.value
  #pval = (1 - pnorm(abs(m[[model]][[1]]$est/m[[model]][[1]]$se.standard))) * 2
  pval = ifelse(pval < 0.001, "0.001", round(pval, 3))
  
  res = data.frame(beta, mean.Tr, mean.Co, pval)
  return(res)
}))
row.names(coefTable) <- namesInput
coefTable$pval.adj <- round(p.adjust(coefTable$pval, method="fdr"),3)

print(kable(coefTable))

	beta	mean.Tr	mean.Co	pval	pval.adj
m3.Ca	-25.622	854.75	880.38	0.156	0.213
m3.Mg	-7.764	204.41	212.18	0.056	0.105
pH	-0.017	5.52	5.54	0.346	0.399
Total.N	-0.002	0.16	0.16	0.065	0.108
Total.C	-0.012	2.11	2.12	0.458	0.491
ExAl	0.026	0.60	0.58	0.189	0.236
n_season_fert	2.547	3.40	0.85	0.001	0.004
n_season_compost	0.354	5.96	5.60	0.004	0.012
n_seasons_leg_1	0.145	2.25	2.11	0.096	0.144
n_season_fallow	0.105	0.67	0.56	0.038	0.081
logFert	0.821	1.19	0.37	0.001	0.004
n_seasons_leg_2	-0.384	2.64	3.02	0.001	0.004
n_season_lime	0.096	0.23	0.13	0.001	0.004
fert1.are	0.009	1.06	1.06	0.947	0.947
compost.are	9.912	54.14	44.23	0.006	0.015

Calculate farmers under soil health thresholds - table 1

thresh <- d %>% group_by(client) %>% dplyr::summarize(
  count = n(),
  ph = sum(pH<5.8),
  carbon = sum(Total.C < 2),
  nitrogen = sum(Total.N < 0.1),
  calcium = sum(m3.Ca < 720),
  magnesium = sum(m3.Mg < 100)
  #aluminum = sum(ExAl)
) %>% mutate(
  under.ph = paste(paste(round(ph/count,4)*100, "%", sep=""), " (", ph, ")", sep=""),
  under.carbon = paste(paste(round(carbon/count,4)*100,"%", sep=""), " (", carbon, ")", sep=""),
  under.nitrogen = paste(paste(round(nitrogen/count,4)*100, "%", sep=""), " (", nitrogen, ")", sep=""),
  under.calcium = paste(paste(round(calcium/count,4)*100, "%", sep=""), " (", calcium, ")", sep=""),
  under.mag = paste(paste(round(magnesium/count,4)*100,"%", sep=""), " (", magnesium, ")", sep="")
) %>% as.data.frame() 

thresh <- thresh[, c("client", names(thresh)[grep("under", names(thresh))])]
thresh <- t(thresh)
colnames(thresh) = thresh[1, ] # the first row will be the header
colnames(thresh) = c("non-client", "client")
thresh = thresh[-1, ]

write.csv(thresh, file=paste(od, "table1_rw_thresholds.csv", sep = "/"), row.names = T)

Table 1:3 coefficients

#
write.csv(coefTable, file=paste(od, "psm coefficients.csv", sep = "/"),
          row.names = T)

# sort by the order Eric wants
t1order <- c("pH", "Total.C", "Total.N", "m3.Ca", "m3.Mg", "ExAl")
table1vars <- paste(t1order, collapse = "|")
table1 <- coefTable[grep(table1vars, rownames(coefTable)), ]
table1 <- table1[order(match(rownames(table1), t1order)),]

write.csv(table1, file=paste(od, "psm coefficients ordered for ES.csv", sep = "/"))

# 11/17 added lime
t2order <- c("n_season_fert", "n_season_compost", "n_season_lime", "n_season_fallow", "n_seasons_leg_1", "n_seasons_leg_2")
table2vars <- paste(t2order, collapse = "|")
table2 <- coefTable[grep(table2vars, rownames(coefTable)), ]
table2 <- table2[order(match(rownames(table2), t2order)),]

write.csv(table2, file=paste(od, "psm coefficients ordered for ES_agprac.csv", sep = "/"))

#table 3
t3order <- c("fert1.are", "compost.are")
table3vars <- paste(t3order, collapse = "|")
table3 <- coefTable[grep(table3vars, rownames(coefTable)), ]
table3 <- table3[order(match(rownames(table3), t3order)),]
write.csv(table3, file=paste(od, "psm coefficients table3.csv", sep = "/"))

Interpretation: Propensity score matching gives us a comparable treatment and control group. The table above shows that after matching on those characteristics, there are effectively no differences between One Acre Fund farmer and Tubura farmers on soil attributes. The unadjusted p-values show 1AF farmers to have slow levels of soil nitrogen but this finding disappears if we account for running multiple matching models.

When we expand the outcome variable set to include practice variables, we first no longer get a good propensity score match for all variables.

• Tubura participation so strongly overlaps with fertilizer use that the comparison group is no longer comparable. The resulting model shows a significant difference in seasons of fertilizer use.
• Tubura farmers have used compost for 0.3 more seasons than non-Tubura farmers. We don’t observe a difference in N or C levels in the soil attributes however.
• We don’t see a difference between Tubura and non-Tubura farmers on seasons of legumes being the primary crop in the study plot.

Study group balance post-match

See here for some guidance on hwo to use weights to reconstruct the group balance following the matches.

See here for weighted t.test documentation

suppressMessages(library(weights))
tableVars <- c("age", "female", "hhsize", "own")

postMatch <- do.call(rbind, lapply(1:length(m), function(model){
  
  innerPost <- do.call(rbind, lapply(tableVars, function(x){
  
  mean.t = weighted.mean(d[m[[model]][[1]]$index.treated,][,x], m[[model]][[1]]$weights)
  mean.c = weighted.mean(d[m[[model]][[1]]$index.control,][,x], m[[model]][[1]]$weights)
  
  # combined data
  dm <- as.data.frame(rbind(d[m[[model]][[1]]$index.treated,], 
              d[m[[model]][[1]]$index.control,]))
  
  test = wtd.t.test(d[m[[model]][[1]]$index.treated,][,x], 
                    d[m[[model]][[1]]$index.control,][,x],
                    weight=m[[model]][[1]]$weights,
                    samedata=TRUE)
  
  return(data.frame(model.num = model, 
                    outcome=x, 
                    tr=mean.t, 
                    contr = mean.c,
                    pval = test$coefficients[3][[1]]))
  }))
  
  return(innerPost)
}))

write.csv(postMatch, file=paste(od, "rw post match covars.csv", sep = "/"),
          row.names = F)

Models with PSM matched farmers

Per Robert’s suggestion, now that we’ve matched Tubura and non-Tubura farmers, let’s assess the severity of Tubura tenure on key soil health outcomes.

tenureTab.add <- lapply(1:length(m), function(model){
  
  dm <- as.data.frame(rbind(d[m[[model]][[1]]$index.treated,], 
              d[m[[model]][[1]]$index.control,]))
  dm$client_tenure <- dm$client*dm$total.seasons
  mod <- lm(as.formula(paste("dm[,psmList[[model]]$y] ~", "total.seasons + as.factor(cell)", sep ="")), data=dm)
  return(mod)
})

# add tenured>=3 to d if we want to run this as a binary comparison

# tenureTab.binary <- lapply(1:length(m), function(model){
#   
#   dm <- as.data.frame(rbind(d[m[[model]][[1]]$index.treated,], 
#               d[m[[model]][[1]]$index.control,]))
#   dm$client_tenure <- dm$client*dm$total.seasons
#   mod <- lm(as.formula(paste("dm[,psmList[[model]]$y] ~", "tenured + as.factor(cell)", sep ="")), data=dm)
#   return(mod)
# })

# modNames <- c("Calcium", "Magnesium", "pH", "Nitrogen", "Carbon", "Seasons fertilizer", "Seasons compost", "Seasons legumes", "Seasons fallow", "Fertilizer (log)", "Season Sec. legumes", "Seasons Lime", "Fertilizer/Are", "Compost/are")

modNames <- unlist(lapply(psmList, function(x){
  return(x$y)
}))

suppressWarnings(
stargazer(tenureTab.add, type="html", 
          title = "2016A Rwanda Soil Health Baseline - PSM Tenure",
          covariate.labels = "One Acre Fund Tenure", 
          dep.var.labels = "",
          #column.labels = modNames,
          #column.labels = c(gsub("m3.", "", as.vector(namesInput))),
          notes = "Includes FE for cell",
          omit=c("cell"), out=paste(od, "rw_baseline_matched_tenure.htm", sep="/"))
)

plm.tenure <- function(x){
  
  intercept = x$coefficients[[1]]
  beta = round(x$coefficients[[2]],3)
  int.pval = summary(x)$coefficients[1,4]
  int.pval = ifelse(int.pval < 0.001, "< 0.001", round(as.numeric(int.pval),3))
  beta.pval = summary(x)$coefficients[2,4]
  beta.pval = ifelse(beta.pval < 0.001, "< 0.001", round(as.numeric(beta.pval),3))
  res = data.frame(intercept, int.pval, beta, beta.pval, stringsAsFactors = F)
  
  return(res)  
}

tenure.reg <- do.call(rbind, lapply(tenureTab.add, function(x){
  plm.tenure(x)
}))

rownames(tenure.reg) <- modNames

t6order <- c("cows", "pigs", "sheep", "goats", "chickens")
table6vars <- paste(t6order, collapse = "|")
rw.table6 <- output[grep(table6vars, rownames(output)), ]
rw.table6 <- rw.table6[order(match(rownames(rw.table6), t4order)),]

ESorder <- c("pH", "Total.C", "Total.N", "m3.Ca", "m3.Mg", "ExAl", "n_season_fert",
             "logFert", "n_season_compost", "n_seasons_leg_1", "n_season_fallow")
ESorder.grep <- paste(ESorder, collapse = "|")
tenure.reg <- tenure.reg[grep(ESorder.grep,rownames(tenure.reg)),]
tenure.reg <- tenure.reg[order(match(row.names(tenure.reg), ESorder)),]

write.csv(tenure.reg, file=paste(od, "table13_reg.csv", sep = "/"), row.names=T)

Interpretation: Using a PSM matched sample, the models above assess the effects of additional years of farming with Tubura. Numerous control farmers have also been Tubura farmers in previous seasons. Thus, I’m keeping the model simple instead of adding a client*tenure interaction. We can easily test that as well though.

Soil Clusters

Heirarchical Clustering

I draw from Robert’s example in the BGMS soil response notebook.

Scale soil variables to remove units as an issue.

scaledSoil <- scale(d[,soilVars])

clustMethods <- c("ward.D2", "single", "complete", "average", "mcquitty", "median", "centroid")
invisible(sapply(clustMethods, FUN=function(method) {
    plot(hclust(dist(scaledSoil), method=method), main=method, ylab="", label=FALSE)
}))

Note: I’m honestly not certain what we’re looking for in these outputs. Look for a balanced tree? Ward looks pretty balanced.

hc <- hclust(dist(scaledSoil), method="ward.D2")

d$hc4 <- cutree(hc, k=4)
hcCentroids <- aggregate(list(d[,soilVars]), by=list(cluster=d$hc4), FUN=mean)
hcTable <- rbind(t(table(d$hc4)),
      t(round(hcCentroids, 2)))

print(kable(hcTable))

	1	2	3	4
	861.00	443.00	922.00	217.00
cluster	1.00	2.00	3.00	4.00
m3.Ca	665.28	380.73	1013.54	2151.18
m3.Mg	169.97	89.04	247.90	434.78
pH	5.28	4.81	5.98	6.22
Total.N	0.16	0.18	0.14	0.19
Total.C	2.11	2.68	1.74	2.59
ExAl	0.64	1.61	0.14	0.11

K-means clustering

K-means clustering is a typical unsupervised learning algorithm. We start by identifying the number of clusters we should tell the formula to look for.

numClusters <- 2:10
plot(numClusters,
     sapply(numClusters, FUN=function(k) { sum(kmeans(scaledSoil, centers=k, nstart=20, iter.max=20)$withinss) }),
     type="b", ylab="average within cluster sum of squared error")

We’re looking for the bend in the graph. It seems to come at 3 but we can try both 3 and 4 and see how they match with our understanding of AEZ and environmental conditions.

set.seed(20161125)
clusters <- 4
d$km4 <- kmeans(scaledSoil, centers=clusters, nstart=20, iter.max=20)$cluster
kmCentroids <- aggregate(list(d[,soilVars]), by=list(cluster=d$km4), FUN=mean)
kmTable <- rbind(t(table(d$km4)),
      t(round(kmCentroids, 2)))

print(kable(kmTable))

	1	2	3	4
	823.00	471.00	673.00	476.00
cluster	1.00	2.00	3.00	4.00
m3.Ca	930.87	1755.64	535.91	397.22
m3.Mg	233.95	371.23	143.52	93.95
pH	5.93	6.12	5.18	4.83
Total.N	0.13	0.18	0.15	0.18
Total.C	1.70	2.41	2.01	2.68
ExAl	0.14	0.12	0.72	1.58

These clusters are roughly equally sized. Robert used another method, partitioning around Mediods, to better account for outliers that were causing the algorithm to create a small additional cluster.

Give the clusters brief descriptions

d$km4 <- factor(d$km4, labels=c("pH=5.9, C=1.7", 
                                "pH=6.1,C=2.4", 
                                "pH=5.1, C=2",
                                "pH=4.8, C=2.6"))

Mappaing the location of the clusters

suppressMessages(library(ggmap))
qmplot(lon, lat, data=d, color=d$km4)

## Using zoom = 9...

## Map from URL : http://tile.stamen.com/toner-lite/9/297/258.png

## Map from URL : http://tile.stamen.com/toner-lite/9/298/258.png

## Map from URL : http://tile.stamen.com/toner-lite/9/299/258.png

## Map from URL : http://tile.stamen.com/toner-lite/9/297/259.png

## Map from URL : http://tile.stamen.com/toner-lite/9/298/259.png

## Map from URL : http://tile.stamen.com/toner-lite/9/299/259.png

## Map from URL : http://tile.stamen.com/toner-lite/9/297/260.png

## Map from URL : http://tile.stamen.com/toner-lite/9/298/260.png

## Map from URL : http://tile.stamen.com/toner-lite/9/299/260.png

## Warning: Removed 84 rows containing missing values (geom_point).

#qmplot(lon, lat, data=d, color=as.factor(hc4))

crdref <- CRS('+proj=longlat +datum=WGS84 +no_defs +ellps=WGS84 +towgs84=0,0,0')
e <- d[!is.na(d$lon),]
ss <- SpatialPointsDataFrame(coords = e[, c("lon", "lat")], data=e, proj4string = crdref)

pal <- colorFactor("RdYlBu", domain=unique(ss$km4))

leaflet() %>% addTiles() %>%
  setView(lng=rwanda$longitude, lat=rwanda$latitude, zoom=8) %>%
  addCircleMarkers(lng=ss$lon, lat=ss$lat, 
                   #radius= ifelse(ss$km4==1, 10,6),
                   color = pal(ss$km4), clusterOptions = markerClusterOptions(disableClusteringAtZoom=11, spiderfyOnMaxZoom=FALSE)) %>% 
  addLegend(pal = pal, values=unique(ss$km4), title = "Clusters ")

Soil clusters by altitude and AEZ

ggplot(d, aes(x=km4, y=alt)) + geom_boxplot()

## Warning: Removed 84 rows containing non-finite values (stat_boxplot).

Soil clusters by altitude and rainfall

Try this when we have aWhere data

Compare clusters to AEZ

print(kable(table(d$km4, d$aez)))

	Central Plateau	Congo Nile	East	Lake Kivu
pH=5.9, C=1.7	299	63	226	235
pH=6.1,C=2.4	70	9	294	98
pH=5.1, C=2	189	200	86	198
pH=4.8, C=2.6	38	327	10	101

We’re not seeing close alignment between clusters and AEZ. This is probably because AEZ are a mix of altitude, rainfall, temperature and soil. When we have weather data we can try to cluster again and see if we’re more closely aligned.

Alternatively, this could be initial evidence that the AEZ designations are not capturing and separating variation very well. I think we need to start with more data though.

Mapping

Soil health maps for soil health study areas

#crdref <- CRS('+proj=longlat +datum=WGS84')
crdref <- CRS('+proj=longlat +datum=WGS84 +no_defs +ellps=WGS84 +towgs84=0,0,0')
e <- d[!is.na(d$lon),]
ss <- SpatialPointsDataFrame(coords = e[, c("lon", "lat")], data=e, proj4string = crdref)

rw <- getData("GADM", country='RW', level=3, path = "/Users/mlowes/drive/optimized_agronomy/soil/soil_health_study/data") # the higher the number, the higher the resolution

#ext.rw.ss matches points with spatial polygons in rw
ext.rw.ss <- extract(rw[, "NAME_3"], ss)

## Loading required namespace: rgeos

ss$spatialname <- ext.rw.ss$NAME_3

frw <- fortify(rw, region="NAME_3")
ss.soil <- aggregate(ss@data[,soilVars], by=list(ss@data$spatialname), function(x){
  mean(x, na.rm=T)
})

plotReady <- dplyr::left_join(frw, ss.soil, by=c("id"="Group.1"))

Plot simple summary of soil characteristics

Consider revising these maps to a smaller greographic unit. Add the name of the location for uninitiated users.

library(RColorBrewer)
mapList <- list()
for(i in 1:length(soilVars)){
mapRes <-  ggplot(plotReady, aes(x=long, y=lat, group=group)) + geom_path() + 
  geom_polygon(aes(fill=plotReady[,soilVars[i]])) + 
  scale_fill_gradientn(colours = rev(brewer.pal(9,"Reds")), # define colors
                       name = soilVars[i],
                       guide = guide_colorbar(legend.direction = "vertical")) + 
  theme_bw() + 
  labs(title=paste("Rwanda long term soil health baseline - 2016", soilVars[i], sep= " "), x = "Longitude", y="Latitude")
mapList[[i]] <- mapRes
print(mapRes)
} 

# This is a small experiment to combine raster (spdf) and leaflet and be able to access the data in the raster interactively.
#mapLayer <- sp::merge(rw, ss.soil, by.x="NAME_3", by.y="Group.1")

# fill = T, fillOpacity = 0.7, fillColor = d.fill, 
#         stroke = T, color = "white", weight = 2, dashArray = 3, 
#         opacity = 0.5, popup = county.tt(d)

# leaflet(mapLayer) %>% addTiles() %>%
#   setView(lng=rwanda$longitude, lat=rwanda$latitude, zoom=8) %>%
#   addPolygons(
#     stroke = TRUE, opacity=0.2, smoothFactor = 0.5, 
#     fillColor=mapLayer$pH, fillOpacity = 0.5)

pdf(file=paste(md, "rw_shs_baseline_soil_maps.pdf", sep = "/"), width=11, height=8.5)
for(i in 1:length(mapList)){
print(mapList[[i]])
}  
dev.off()

## quartz_off_screen 
##                 2

Interpolations of soil health values

Interpolate soil health values for full operating area using soil health study values. We want to eventually add all Rwandan soil values into a single dataset to update and hone these values. See here for more guidanace

note:

• Layer on the Tubura sites to the interpolated map.
• Use leaflet so you can zoom in and out more easily.
• Make it a raster layer that you can click on understand the values at different locations and the name of the location. I’ll need to make this a shiny app to have that functionality

The code below will run 5 K-fold cross validation to compare interpolation models. The output will be fed into the interpolate leaflet code below.

Check that I’m handling the projection correctly with Robert

# proj4string(ss) <- CRS("+init=epsg:4326")
# ss <- spTransform(ss, CRS=("+proj=utm +zone=36N +datum=WGS84"))

# root mean sq error for evaluating models
RMSE <- function(observed, predicted) {
  sqrt(mean((predicted - observed)^2, na.rm=TRUE))
}

# set k folds to 5
set.seed(20161030)
nfolds <- 5
k <- kfold(ss, nfolds) # from dismo

# cross validation of models
ensrmse <- tpsrmse <- idwrmse <- rep(NA, 5) # assing multiple objects at once

cv <- function(x) {
  
  for(i in 1:nfolds) {
    train <- ss[k!=i,]
    test <- ss[k==i,]

  train <- train[!is.na(train@data[,x]),]
  
  # inverse distance weights
  m <- gstat(formula=as.formula(paste(x, '~ 1')), locations=train)
  p1 <- predict(m, newdata=test, debug.level=0)$var1.pred
  idwrmse[i] <-  RMSE(test@data[,x], p1) #idw rsme
  
  # krieging
  # m <- autoKrige(formula=as.formula(paste(x, "~ 1")), input_data = train)
  # p2 <- predict(m, newdata=test, debug.level=0)$var1.pred
  # krigrmse[i] <-  RMSE(test$OZDLYAV, p2)
  
  # thin plate spline
  m <- Tps(coordinates(train), train@data[,x]) 
  p3 <- predict(m, coordinates(test))
  tpsrmse[i] <-  RMSE(test@data[,x], p3)
  
  w <- c(idwrmse[i], tpsrmse[i]) # combine the rmse
  weights <- w / sum(w) # weight them
  ensemble <- p1 * weights[1] + p3 * weights[2] 
  # multiply predictions by weights
  ensrmse[i] <-  RMSE(test@data[,x], ensemble) # truly an ensemble result?
  }
  
  output <- rbind(idwrmse, tpsrmse, ensrmse)
  return(output)
  
}

Now loop over the variables of interest where x is the soilVar variable.

output <- lapply(soilVars, function(x){
  ini <- data.frame(cv(x))
  ini$ave <- apply(ini[,1:5], 1, function(y){mean(y, na.rm=T)})
  res <- paste("Best model is ", row.names(ini[which.min(ini$ave),]),  sep = "")
  return(list(ini, res))
})

Interpolated soil maps

r <- raster(res=1/12)
r <- crop(r, floor(extent(rw)))

maps <- lapply(soilVars, function(x){
  m <- Tps(coordinates(ss), ss@data[,x])
  # make raster layer with model, raster is rwanda empty raster, model is m
  tps <- interpolate(r, m)
  tps <- crop(tps, rw)
  tps <- raster::mask(tps, rw) # cuts the tps raster down to the rw boundaries
  x <- gsub("m3.", "", x)
  
  palColors <- leaflet::colorNumeric(palette = "Reds", values(tps), na.color = "transparent")
  
  suppressWarnings(
  leaflet() %>% addTiles() %>%
  addRasterImage(tps, colors=palColors, opacity = 0.8) %>%
  setView(lng=rwanda$longitude, lat=rwanda$latitude, zoom=8) %>%
  addLegend(pal = palColors, values = values(tps), title = paste("Soil Value ", x, sep=""))
  )
})

save(maps,ss, file=paste(dd, "rw_baseline_interpolation_maps.Rdata", sep = "/"))

tagList(maps)

Print out the interpolated values for inclusion in the report.

pdf(file=paste(md, "rw_shs_bl_interpolation_soil_vars.pdf", sep = '/'), width=11, height=8.5)
lapply(soilVars, function(x) {
  m <- Tps(coordinates(ss), ss@data[,x])
  # make raster layer with model, raster is rwanda empty raster, model is m
  tps <- interpolate(r, m)
  tps <- crop(tps, rw)
  tps <- raster::mask(tps, rw) # cuts the tps raster down to the rw boundaries
  x <- gsub("m3.", "", x)
  
suppressMessages(
  outPlot <- invisible(print(
  plot(tps, main= paste("Soil TPS Interpolation ", x, sep="- ")),
  plot(rw, add=T, na.print=NULL)
  ))
)
})

## NULL

## NULL

## NULL

## NULL

## NULL

## NULL

## [[1]]
## NULL
## 
## [[2]]
## NULL
## 
## [[3]]
## NULL
## 
## [[4]]
## NULL
## 
## [[5]]
## NULL
## 
## [[6]]
## NULL

dev.off()

## quartz_off_screen 
##                 2

–end